Medical Aspects of Chemical Warfare

Home , GH (nerve agent), GP (nerve agent), Perfluoroisobutene

Medical Diagnostics

Chapter 22 MEDICAL DIAGNOSTICS

† ‡ § Benedict R. Capacio, Phd*; J. Richard Smith ; Richard K. Gordon, Phd ; Julian R. Haigh, Phd ; John ¥ ¶ R. Barr, Phd ; a n d Gennady E. Platoff Jr, Phd

INTRODUCTION

NERVE AGENTS

SULFUR MUSTARD

LEWISITE

CYANIDE

PHOSGENE

3-QUINUCLIDINYL BENZILATE

SAMPLE CONSIDERATIONS

Summary

* Chief, Medical Diagnostic and Chemical Branch, Analytical Toxicology Division, US Army Medical Research Institute of Chemical Defense, 3100 Rickets Point Road, Aberdeen Proving Ground, Maryland 21010-5400 † Chemist, Medical Diagnostic and Chemical Branch, Analytical Toxicology Division, US Army Medical Research Institute of Chemical Defense, 3100 Rickets Point Road, Aberdeen Proving Ground, Maryland 21010-5400 ‡ Chief, Department of Biochemical Pharmacology, Biochemistry Division, Walter Reed Army Institute of Research, 503 Robert Grant Road, Silver Spring, Maryland 20910-7500 § Research Scientist, Department of Biochemical Pharmacology, Biochemistry Division, Walter Reed Army Institute of Research, 503 Robert Grant Road, Silver Spring, Maryland 20910-7500 ¥ Lead Research Chemist, Centers for Disease Control and Prevention, 4770 Buford Highway, Mailstop F47, Atlanta, Georgia 30341 ¶ Colonel, US Army (Retired); Scientific Advisor, Office of Biodefense Research, National Institute of Allergies and Infectious Disease, National Institutes of Health, 6610 Rockledge Drive, Room 4069, Bethesda, Maryland 20892-6612

691 Medical Aspects of Chemical Warfare

INTRODUCTION

In the past, issues associated with chemical war- an exposed individual may be considered as a potential fare agents, including developing and implementing matrix (eg, blister fluid from sulfur mustard [North medical countermeasures, field detection, verifica- Atlantic Treaty Organization (NATO) designation: tion of human exposures, triage, and treatment, have HD] vesication). Regardless, analyses of these types of primarily been a concern of the military community samples are inherently difficult because of the matrix’s because most prior experience with chemical war- complex composition and the presence of analyte in fare agents was limited to the battlefield. However, trace quantities. chemical agents have been increasingly employed Noninvasive urine collection does not require highly against civilian populations, such as in Iraqi attacks trained medical personnel or specialized equipment. against the Kurds and the attacks organized by the Although biomarkers present in urine are usually short- Aum Shinrikyo cult in Matsumoto City and the Tokyo lived (hours to days) metabolites, they can be present subway. The attacks on the World Trade Center in in relatively high concentrations in samples obtained New York City and the Pentagon in Washington, DC, shortly after exposure. The collection of blood/plasma in 2001 have increased concern about the potential should be performed by trained medical personnel. large-scale use of chemical warfare agents in a civilian Blood/plasma samples offer potential benefits because sector. Incidents involving large numbers of civilians both metabolites and the more long-lived adducts of a have shown that to facilitate appropriate treatment, it macromolecular target can be assayed. The effective- is critical to identify not only those exposed, but those ness of using tissue to verify chemical agent exposure is who have not been exposed, as well. In addition to generally limited to postmortem sampling. For example, health issues associated with exposure, the political formalin-fixed brain tissues from fatalities of the Tokyo and legal ramifications of a chemical warfare attack subway attack were successfully used to verify sarin can be enormous. It is therefore essential that testing (NATO designation: GB) as the agent employed in that for exposure be accurate, sensitive, and rapid. attack.1 Other tissue samples can be obtained from the For the most part, monitoring for the presence of carcasses of animals at the incident site. chemical warfare agents in humans, or “biomonitor- Methods that do not directly analyze cholinesterase ing,” involves examining specimens to determine if (ChE) activity typically involve detection systems like an exposure has occurred. Assays that provide de- mass spectrometry (MS) combined with either gas finitive evidence of agent exposure commonly target chromatography (GC) or liquid chromatography (LC) metabolites, such as hydrolysis products and adducts to separate the analyte from other matrix components. formed following binding to biomolecular entities. MS detection methods are based upon specific and Unlike drug efficacy studies in which blood/plasma characteristic fragmentation patterns of the parent levels usually focus on the parent compounds, assay molecule, making MS detection desirable because it techniques for verifying chemical agent exposure identifies the analyte fairly reliably. Other detection rarely target the intact agent because of its limited systems, such as nitrogen-phosphorus detection and longevity in vivo. Following exposure, many agents flame photometric detection, have also been used. are rapidly converted and appear in the blood as hy- Some analytes can be directly assessed, whereas others drolysis products (resulting from reactions with water) may require chemical modification (eg, derivatization) and are excreted in urine. Because of rapid formation to enhance detection or make them more volatile in and subsequent urinary excretion, the use of these GC separations. More sophisticated techniques may products as markers provides a limited window of op- employ GC or LC with tandem MS (MS-MS) detection portunity to collect a sample with measurable product. systems, allowing more sensitivity and selectivity. More long-lived markers tend to be those that result Validating the performance of an analytical tech- from agent interactions with large-molecular–weight nique subsequent to initial in-vitro method develop- targets, such as proteins and DNA. These result from ment is usually accomplished with in-vivo animal covalent binding of the agent or agent moiety to form exposure models. Additional information may be macromolecular adducts. As such, the protein acts as gleaned from archived human samples from past a depot for the adducted agent and the residence time exposure incidents. Samples from humans exposed to is similar to the half-life of the target molecule. sulfur mustard during the Iran-Iraq War and a limited Blood/blood components, urine, and, in rare in- number of samples from the Japanese nerve agent at- stances, tissue specimens, termed “sample matrices” or tacks have been used to evaluate assay techniques. In “biomedical samples,” can be used to verify chemical the case of some agents, background marker levels are agent exposure. However, any sample obtained from known to exist in nonexposed individuals, making it

692 Medical Diagnostics

difficult to interpret the results of potential incidents. Preexposure treatments or tests to monitor potential Therefore, in addition to assessing performance in ani- chemical agent exposure may be warranted for military mal models and archived human samples, it is essential personnel and first responders who must enter or oper- to determine potential background levels and inci- ate in chemically contaminated environments. How- dence of markers in nonexposed human populations. ever, laboratory testing may not be as useful for large It should be stressed that, for the purpose of exposure civilian populations unless there is a clear impending verification, results from laboratory testing must be chemical threat. At the same time, determining the considered along with other information, such as the health effects of chemical exposure is complex because presentation of symptoms consistent with the agent in it can affect the nervous system, respiratory tract, skin, question and results from environmental testing. eyes, and mucous membranes, as well as the gastroin- In the late 1980s the US Army Medical Research testinal, cardiovascular, endocrine, and reproductive Institute of Chemical Defense was tasked by the De- systems. Individual susceptibility, preexisting medical partment of Defense to develop methods that could conditions, and age may also contribute to the severity confirm potential chemical warfare agent exposure. of a chemically related illness. Chronic exposures, even The US Army Medical Research Institute of Chemi- at low concentrations, are another concern. In addition cal Defense had previously published procedures for to development of diagnostic technologies, strategies verifying exposure to nerve agents and sulfur mustard. to detect chemical agent exposure have become a These methods primarily focused on GC-MS analysis public health issue. of hydrolysis products excreted in the urine following The transition of laboratory-based analytical tech- exposure to chemical warfare agents. Subsequently, the niques to a far-forward field setting can generate methods using urine or blood samples were compiled valuable information for military or civilian clinicians. as part of Technical Bulletin Medical 296, titled “Assay In this transition, problems such as data analysis, Techniques for Detection of Exposure to Sulfur Mus- interpreting complex spectra, and instrument trouble- tard, Cholinesterase Inhibitors, Sarin, Soman, GF, and shooting and repair may need addressing. As analyti- Cyanide.”2 The publication was intended to provide cal methods are developed, refined, and sent farther clinicians with laboratory tests to detect exposure to from the laboratory, advanced telecommunication will chemical warfare agents. be needed to provide a direct link between research In the mid 1990s, after the publication of Technical scientists and field operators; telecommunication will Bulletin Medical 296, the military adapted some of the become critical to confirming patient exposure and laboratory analytical methods for field-forward use. tracking patient recovery and treatment. The concept was demonstrated by the US Army 520th This chapter provides a basic outline and refer- Theater Army Medical Laboratory, which used the ences for state-of-the-art analytical methods presently Test-Mate OP Kit (EQM Research Inc, Cincinnati, OH) described in the literature. Methods of verification for acetylcholinesterase (AChE) assay and a fly-away for exposure to nerve agents, vesicants, pulmonary GC-MS system. The lengthy preparation of GC and MS toxicants, metabolic poisons, incapacitating agents, samples for analysis in a field environment was one and riot control agents will be reviewed. Biological of the reasons that alternative methods of analysis for sample collection, handling, storage, shipping, and chemical warfare agents were later examined. submission will be explained.

NERVE AGENTS

Background amounts were produced by the end of the war.4 Five of the OP compounds are generally regarded as nerve The first organophosphorus (OP) nerve agent, agents: tabun, sarin, soman, cyclosarin (NATO des- tabun (NATO designation: GA) was developed ignation: GF), and Russian VX.4 These compounds shortly before and during World War II by German demonstrated extreme toxicity, which was attributed chemist Gerhard Schrader at IG Farbenindustrie in to long-lasting binding and inhibition of the enzyme an attempt to develop a commercial insecticide.3,4,5 AChE. As a result, the compounds were referred to Shortly thereafter, sarin was synthesized. Both are as “irreversible” inhibitors. Related, but less toxic extremely toxic. The German government realized compounds (ie, “reversible” inhibitors), are becoming the compounds had potential as chemical warfare widely used therapeutically; for example, in the treat- agents and began producing them and incorporating ment of Alzheimer’s disease. The relative description them into munitions. Subsequently, soman (NATO as reversible or irreversible refers to the length of the designation: GD) was synthesized, but only small binding to the enzyme (Figure 22-1).

693 Medical Aspects of Chemical Warfare

Sarin (GB) Soman (GD) Cyclosarin (GF) O O O

H3C P OCH(CH3)2 H3C P OCHC(CH3)3 H3C P O

F F CH3 F

Tabun (GA) VX Russian VX O O O

NC P OCH2CH3 H3C P OCH2CH3 H3C P OCH2CH(CH3)2

N(CH3)2 SCH2CH2N CH(CH3)2 SCH2CH2N CH3

CH(CH3)2 CH3

Fig. 22-1. Chemical structures of nerve agents. The nerve agents sarin (GB), soman (GD), and cyclosarin (GF) lose fluorine subsequent to binding to cholinesterase. The agents tabun (GA), VX, and Russian VX lose cyanide and the thiol groups.

Many of the assays developed for exposure verifica- lab-based, non-ChE analytical methods have been tion are based on the interaction of nerve agents with developed, and several successfully utilized, to verify ChE enzymes. Nerve agents inhibit ChE by forming nerve agent exposure. For the most part, these employ a covalent bond between the phosphorus atom of the MS with GC or LC separations. The tests are relatively agent and the serine residue of the enzyme active site. labor intensive, requiring trained personnel and so- That interaction results in the displacement or loss of phisticated instrumentation not usually available in fluorine from sarin, soman, and cyclosarin. The bind- clinical settings. Most experience using these tech- ing of tabun, VX, and Russian VX is different in that the niques has come from animal exposure models. These leaving group is cyanide followed by the thiol groups assessments allow for determination of test sensitivity (see Figure 22-1).6 Spontaneous reactivation of the and biomarker longevity in experimental models. In enzyme or hydrolysis reactions with water can occur humans, there is limited experience from accidental to produce corresponding alkyl methylphosphonic and terror-related exposures. This chapter will review acids (MPAs). Alternatively, the loss of the O-alkyl assays for chemical warfare agent exposure that have group while bound to the enzyme produces a highly been published in the literature and how they have stable organophosphoryl-ChE bond, a process referred been applied in potential exposure situations to as “aging.” Once aging has occurred, the enzyme is considered resistant to reactivation by oximes or other Assay of Parent Compounds nucleophilic reagents.4 The spontaneous reactivation and aging rates of the agents vary depending on the O- Analyzing for parent nerve agents from biomedi- alkyl group. For example, VX-inhibited red blood cell cal matrices, such as blood or urine, is not a viable (RBC) ChE reactivates at an approximate rate of 0.5% diagnostic technique for retrospective detection of to 1% per hour for the first 48 hours, with minimal ag- exposure.7 Parent agents are relatively short-lived being. On the other hand, soman-inhibited ChE does not cause of rapid hydrolysis and binding to plasma and spontaneously reactivate and has a very rapid aging tissue proteins, imposing unrealistic time restraints rate, with a half-time of approximately 2 minutes.4 on sample collection. The short residence time is especially profound with the G agents (relative to VX). General Clinical Tests Following the intravenous administration of soman at

2 times the median lethal dose (LD50) results in parent With the exception of ChE analysis, there are no agent detection at toxicologically relevant levels for standard clinical assays that specifically test for nerve 104 and 49 minutes in guinea pigs and marmosets, reagent exposure. However, over the years numerous spectively; rapid elimination was reflected in terminal

694 Medical Diagnostics

half-life rates (16.5 min for guinea pigs; 9 min for and VX, respectively (Figure 22-2). Additionally for marmosets).8 Inhalation experiments using nose-only VX, hydrolysis of the sulfur-phosphorus bond occurs,

exposure of guinea pigs to 0.8 × LCt50 (the vapor or yielding DAET and EMPA. The formation and assay aerosol exposure that is lethal to 50% of the exposed of DAET has been reported in rat plasma spiked with population) agent demonstrate terminal half-lives of VX.11 Furthermore, the presence of diisopropyl amino- approximately 36 and 9 minutes for sarin and soman, ethyl methyl sulfide, presumably resulting from the respectively.9 In contrast, similar studies with VX in in-vivo methylation of DAET, has been reported in hu- hairless guinea pigs and marmosets indicate VX is man exposures.18 To date, numerous variations of the more persistent than the G agents.10 These studies alkyl MPA assay for biological fluids, such as plasma show that VX can be found at acutely toxic levels for and urine, have been developed. These include GC 10 to 20 hours following intravenous administration separations with MS,18–20 tandem MS (MS-MS),18,19,21,22 23,24 at a dose one or two times the LD50, with terminal and flame photometric detection. Other methods 25 elimination rates of 98 minutes (1 times the LD50 in involving LC with MS-MS and indirect photometric 26 hairless guinea pigs), 165 minutes (2 times the LD50), detection have also been reported (Table 22-1). and 111 minutes in marmosets (at a dose equivalent to 1 LD50 in hairless guinea pigs). Percutaneous ad- Application to Human Exposures ministration of the LD50 of VX to hairless guinea pigs demonstrated relatively low blood levels (140 pg/ The utility of some methodologies has been dem- mL), which reached a maximum after approximately onstrated in actual human exposure incidents. Most 6 hours.10 Because the route of human exposure to involve assays of urine and plasma or serum. Tsuchi- VX would most likely occur percutaneously, the time hashi et al18 demonstrated the presence of EMPA in frame of 6 hours may be the more relevant assessment the serum of an individual assassinated with VX in of its persistence in blood. This allows very limited Osaka, Japan, in 1994. As mentioned earlier, these time for sample collection and analysis. Others have authors also reported the presence of diisopropyl demonstrated that VX can be assayed from spiked aminoethyl methyl sulfide, which resulted from the rat plasma.11 These authors noted that 53% of the VX in-vivo methylation of DAET subsequent to cleavage of was lost in spiked plasma specimens after 2 hours. the sulfur-phosphorus bond. Reported concentrations The disappearance was attributed to the enzyme in serum collected 1 hour after exposure were 143 ng/ action of the OP hydrolase splitting or to cleavage mL diisopropyl aminoethyl methyl sulfide and 1.25 of the sulfur-phosphorus bond to form diisopropyl μg/mL for EMPA. aminoethanethiol (DAET) and ethyl methylphospho- The Aum Shinrikyo cult attacked citizens twice in nic acid (EMPA).11 Japan using sarin. The first was in an apartment complex in Matsumoto City, where approximately 12 liters Assay of Hydrolysis Compounds of sarin were released using a heater and fan. Accord- ing to police reports, 600 inhabitants in the surround- Analytical Methods ing area were harmed, including 7 who were killed. In the second attack, sarin was released into the Tokyo An alternative approach to direct assay of parent subway, resulting in more than 5,000 casualties and 10 nerve agents is to measure metabolic or hydrolysis deaths.27 Assay of hydrolysis products as a definitive products in specimens. These compounds are pro- marker were used to verify that sarin was the agent em- duced in vivo as a result of hydrolysis or detachment ployed in these events. Minami et al23 and Nakajima et following spontaneous regeneration of the AChE al24 demonstrated the presence of IMPA or MPA in vic- enzyme. Studies of parent nerve agents with radioiso- tims’ urine following sarin exposure in the Tokyo and topically labeled phosphorus (32P) or hydrogen (3H) in Matsumoto attacks, respectively. These methods used animals suggest that agents are rapidly metabolized GC separations of the prepared urine matrix coupled and hydrolyzed in the blood and appear in the urine with flame photometric detection. In the Matsumoto as their respective alkyl MPAs.12–15 This observation incident, urinary concentrations of IMPA and MPA, led to the development of assays for alkyl MPAs in as well as the total dose of the sarin exposure, were biological samples,16 the applicability of which was reported.24 For one victim, MPA concentrations were subsequently demonstrated in animals exposed to 0.14 and 0.02 ug/mL on the first and third days after nerve agents.17 The common products found are exposure, and 0.76, 0.08, and 0.01 ug/mL for IMPA, isopropyl methylphosphonic acid (IMPA), pinacolyl respectively, on the first, third, and seventh days after methylphosphonic acid, cyclohexyl methylphosphonic exposure.24 In this case, the individual was estimated acid, and EMPA derived from sarin, soman, cyclosarin, to have been exposed to 2.79 mg of sarin.

695 Medical Aspects of Chemical Warfare

Sarin (GB) Soman (GD) Cyclosarin (GF) O O O

H3C P OCH(CH3)2 H3C P OCHC(CH3)3 H3C P O

F F CH3 F

O O O

H3C P OCH(CH3)2 H3C P OCHC(CH3)3 H3C P O

OH OH CH3 OH

IMPA PMPA CMPA

H3C P OH

MPA

Fig. 22-2. Hydrolysis pathway of sarin (GB), soman (GD), and cyclosarin (GF). Hydrolysis pathway of nerve agents proceeds through the alkyl methylphosphonic acids IMPA, PMPA, and CMPA to MPA. Analysis of the alkyl methylphosphonic acids allows identification of the parent agent, while assay of MPA is nonspecific. CMPA: cyclohexyl methylphosphonic acid IMPA: isopropyl methylphosphonic acid MPA: methylphosphonic acid PMPA: pinacolyl methylphosphonic acid

Although the report on the Tokyo23 incident did not dure. Reported serum concentrations ranged from 2 directly indicate urinary concentrations, total sarin to 127 ng/mL and 2 to 135 ng/mL in the Tokyo and exposure was estimated. The exposure estimates in a Matsumoto incidents, respectively.25,28 Samples were comatose individual was 0.13 to 0.25 mg/person and obtained 1.5 hours after the incident. In some cases a 0.016 to 0.032 mg/person in a less severely exposed second sample was obtained 2 to 2.5 hours after the casualty.23 These numbers are approximately 10-fold incident; in those samples, the authors report signifi- less than those reported by Nakajima et al24 for a se- cantly lower IMPA concentrations consistent with the verely intoxicated patient. Consistent with rapid elimi- rapid elimination of these compounds. The sarin dose nation, the maximum urinary concentration of these in both incidences was calculated to be 0.2 to 15 mg/ compounds was reported to have occurred in 12 hours person.25 These reported values for sarin exposure are of exposure. Using LC-MS-MS methods, Noort et al25 in the range of those reported by Nakajima et al24 and and Polhuijs et al28 detected IMPA in serum samples are approximately 10-fold greater than those reported from both the Tokyo and Matsumoto incidents. This by Minami et al.23 assay involves fairly sophisticated instrumentation, Alkyl MPAs provide a convenient marker for but allows for a simplified sample processing proce- determining exposure to nerve agents. Numerous

696 Medical Diagnostics

Table 22-1 Analytical Methods for Assay of Nerve Agent Hydrolysis Products*

Sample Matrix Product Identified Analytical Method

Blood, Plasma, Urine, Lung Tissue impa, CMPA, PMPA GC-MS1,2 Serum, Urine EMPA, IMPA, PMPA GC-MS, GC-MS-MS3 Plasma DAET GC-MS4 Urine EMPA, IMPA, MPA GC-FPD5 Urine IMPA, MPA GC-FPD6 Serum EMPA, DAEMS GC-MS, GC-MS-MS7 Serum, Urine, Saliva EMPA, IMPA, PMPA GC-MS8 Urine EMPA, IMPA, CMPA, PMPA, GA acid GC-MS-MS9 Urine IMPA LC-MS-MS10,11 Serum EMPA, IMPA, MPA PMPA Indirect Photometric Detection Ion Chromatography12 Urine, Saliva EMPA, IMPA, CMPA, MPA, pmpa lc-MS-MS13 Urine EMPA, RVX acid, IMPA, PMPA, CMPA, GC-MS-MS14 GA acid, GA diacid

* Although the sample matrices and analytical methods for some of the assays are similar, the authors specifically identified the products listed. CMPA: cyclohexyl methylphosphonic acid DAEMS: diisopropyl aminoethyl methyl sulfide (resulting from the metabolic methylation of DAET) DAET: diisopropyl aminoethanethiol EMPA: ethyl methylphosphonic acid FPD: flame photometric detection GA: tabun GC: gas chromatography IMPA: isopropyl methylphosphonic acid LC: liquid chromatography MPA: methylphosphonic acid MS: mass spectrometry PMPA: pinacolyl methylphosphonic acid RVX: Russian VX Data sources: (1) Shih ML, Smith JR, McMonagle JD, Dolzine TW, Gresham VC. Detection of metabolites of toxic alkylmethylphosphonates in biological samples. Biol Mass Spectrom. 1991;20:717–723. (2) Shih ML, McMonagle JD, Dolzine TW, Gresham VC. Metabolite pharmacokinetics of soman, sarin, and GF in rats and biological monitoring of exposure to toxic organophosphorus agents. J Appl Toxicol. 1994:14:195–199. (3) Fredriksson SA, Hammarström LG, Henriksson L, Lakso HA. Trace determination of alkyl methylphosphonic acids in environmental and biological samples using gas chromatography/negative-ion chemical ionization mass spectrometry and tandem mass spectrometry. J Mass Spectrom. 1995;30:1133–1143. (4) Bonierbale E, Debordes L, Coppet L. Application of capillary gas chromatography to the study of hydrolysis of the nerve agent VX in rat plasma. J Chromatogr B Biomed Sci Appl. 1997;688:255–264. (5) Minami M, Hui DM, Katsumata M, Inagaki H, Boulet CA. Method for the analysis of methylphosphonic acid metabolites of sarin and its ethanol-substituted analogue in urine as applied to the victims of the Tokyo sarin disaster. J Chromatogr B Biomed Sci Appl. 1997;695:237–244. (6) Nakajima T, Sasaki K, Ozawa H, Sekjima Y, Morita H, Fukushima Y, Yanagisawa N. Urinary metabolites of sarin in a patient of the Matsumoto incident. Arch Toxicol. 1998;72:601–603. (7) Tsuchihashi H, Katagi M, Nishikawa M, Tatsuno M. Identification of metabolites of nerve agent VX in serum collected from a victim. J Anal Toxicol. 1998;22:383–388. (8) Miki A, Katagi M, Tsuchihashi H, Yamashita M. Determination of alkylmethylphosphonic acids, the main metabolites of organophosphorus nerve agents, in biofluids by gas chromatography-mass spectrometry and liquid-liquid-solid-phase- transfer-catalyzed pentafluorobenzylation. J Anal Toxicol. 1999;23:86–93. (9) Driskell WJ, Shih M, Needham LL, Barr DB. Quantitation of organophosphorus nerve agent metabolites in human urine using isotope dilution gas chromatography-tandem mass spectrometry. J Anal Toxicol. 2002;26:6–10. (10) Noort D, Hulst AG, Platenburg DH, Polhuijs M, Benschop H. Quantitative analysis of O-isopropyl methylphosphonic acid in serum samples of Japanese citizens allegedly exposed to sarin: estimation of internal dosage. Arch Toxicol. 1998;72:671–675. (11) Polhuijs M, Langenberg JP, Noort D, Hulst AG, Benschop HP. Retrospective detection of exposure to organophosphates: analyses in blood of human beings and rhesus monkeys. In: Sohns T, Voicu VA, eds. NBC Risks: Current Capabilities and Future Perspectives for Protection. Dordrecht, Holland, Netherlands: Kluwer Academic Publishers; 1999:513–521. (12) Katagi M, Nishikawa M, Tatsuno M, Tsuchihashi H. Determination of the main hydrolysis products of organophosphorus nerve agents, methylphosphonic acids, in human serum by indirect photometric detection ion chromatography. J Chromatogr B Biomed Sci Appl. 1997;698:81–88. (13) Hayes TL, Kenny DV, Hernon-Kenny L. Feasibility of direct analysis of saliva and urine for phosphonic acids and thiodiglycol-related species associated with exposure to chemical warfare agents using LC-MS/MS. J Med Chem Def. 2004;2:1-23. (14) Barr JR, Driskell WJ, Aston LS, Martinez RA. Quantitation of metabolites of the nerve agents sarin, soman, cyclosarin, VX, and Russian VX in human urine using isotope-dilution gas chromatography-tandem mass spectrometry. J Anal Toxicol. 2004;28:371–378.

697 Medical Aspects of Chemical Warfare modifications of the assay for these compounds have and biological targets with large molecular weights been developed, and several have been applied to (adducts to biomolecules), such as proteins. The reac- human exposure cases. Important factors to consider tion of chemical agents with large molecules provides when anticipating using this test are the extent of a pool of bound compound that can be tested to verify exposure and time elapsed since the event. In most exposure. Theoretically, the longevity of the marker cases, hydrolysis products are not expected to be pres- is consistent with the in-vivo half-life of the target ent for more than 24 to 48 hours following exposure; molecule, provided that the binding affinity is high however, one of the most severely poisoned victims of enough to prevent spontaneous reactivation. Binding the Matsumoto sarin attack had measurable IMPA in of nerve agents to ChE targets has been one of the the urine on the seventh day after the incident. In this primary interactions leveraged in assay development. particular case, extremely depressed AChE values, in Several assays have been developed based on varia- range of 5% to 8% of normal,24 further indicated the tions of this concept. extent of exposure (Table 22-2). Analytical Methods Assay of Adducts to Biomolecules Polhuijs et al29 developed an assay technique based The relatively rapid excretion and short-lived pres- on observations of earlier findings that sarin-inhibited ence of urinary hydrolysis products imposes time ChE could be reactivated with fluoride ions.30–32 The restrictions for collecting a viable sample. Efforts to displacement of covalently bound sarin to butyrylcho- increase the sampling window have taken advantage linesterase (BChE) was accomplished by incubating of the interactions between chemical warfare agents inhibited plasma with fluoride to form free enzyme

Table 22-2 Methods Used to Confirm Human Exposures to Nerve Agents via Assay of Hydrolysis Products

Agent/Incident Sample Matrix Product Identified Concentration Reported Analytical Method

GB, Tokyo, Japan Urine EMPA, IMPA, mpa nR GC-FPD1 GB, Matsumoto, Japan urine IMPA 0.76–0 .01 μg/mL GC-FPD2 MPA 0.14–0.02 μg/mL GB, Matsumoto and Tokyo, Serum IMPA Matsumoto (2–135 ng/mL) lc-MS-MS3,4 Japan Tokyo (2–127 ng/mL) VX, Osaka, Japan Serum EMPA, diisopro- 1.25 μg/mL GC-MS, GC-MS-MS5 pylaminoethyl 143 ng/mL methyl sulfide

EMPA: ethyl methylphosphonic acid FPD: flame photometric detection GB: sarin GC: gas chromatography IMPA: isopropyl methylphosphonic acid LC: liquid chromatography MPA: methylphosphonic acid MS: mass spectrometry NR: not reported Data sources: (1) Minami M, Hui DM, Katsumata M, Inagaki H, Boulet CA. Method for the analysis of methylphosphonic acid metabolites of sarin and its ethanol-substituted analogue in urine as applied to the victims of the Tokyo sarin disaster. J Chromatogr B Biomed Sci Appl. 1997;695:237–244. (2) Nakajima T, Sasaki K, Ozawa H, Sekjima Y, Morita H, Fukushima Y, Yanagisawa N. Urinary metabolites of sarin in a patient of the Matsumoto incident. Arch Toxicol. 1998;72:601–603. (3) Noort D, Hulst AG, Platenburg DH, Polhuijs M, Benschop H. Quantita- tive analysis of O-isopropyl methylphosphonic acid in serum samples of Japanese citizens allegedly exposed to sarin: estimation of internal dosage. Arch Toxicol. 1998;72:671–675. (4) Polhuijs M, Langenberg JP, Noort D, Hulst AG, Benschop HP. Retrospective detection of exposure to organophosphates: analyses in blood of human beings and rhesus monkeys. In: Sohns T, Voicu VA, eds. NBC Risks: Current Capabilities and Future Perspectives for Protection. Dordrecht, Holland, Netherlands: Kluwer Academic Publishers; 1999:513–521. (5) Tsuchihashi H, Katagi M, Nishikawa M, Tatsuno M. Identification of metabolites of nerve agent VX in serum collected from a victim. J Anal Toxicol. 1998;22:383–388.

698 Medical Diagnostics

plus the parent agent (isopropyl methylphosphono- kaline phosphatase digestion process. The analytical fluoridate). Following isolation from the matrix with technique used is similar to numerous other GC-MS solid phase extraction techniques, the agent was then assays for hydrolysis products. analyzed using GC with MS or other appropriate Another approach based on OP binding to BChE detection systems. Other research has demonstrated has been reported by Fidder et al.44 This method conceptually similar approaches for detecting tabun28 involves digesting BChE to produce nonapeptide and VX.33 In the case of tabun, the cyanide group, fragments containing the serine-198 residue to which which is initially lost upon binding to the enzyme, nerve agents bind. Analyzing nonapeptides employs is replaced with fluorine, leading to the formation of LC-MS-MS techniques. The utility of this method was O-ethyl N,N-dimethyl-phosphoramidofluoridate, a demonstrated by analyzing two archived samples fluorinated analog of tabun.28 Similarly, the thiol group from the Tokyo subway terrorist attack. The authors in VX, which is initially lost upon binding to the en- reported results similar to a previous analysis of those zyme, is replaced by fluorine, resulting in a fluorinated samples, in this case using the fluoride regeneration analog of VX (ethyl methyl-phosphonofluoridate; procedure.44 A reported advantage of this technique VX-G).33 Variations and improvements of the fluoride is that aged or nonaged OPs can be successfully regeneration procedure have evolved to enhance test identified. In the case of sarin-inhibited enzyme, the sensitivity by optimizing agent extraction, increasing serine-198 is conjugated to IMPA; for soman following injection volumes (thermal desorption and large vol- loss of the pinacolyl alkyl group (ie, aging), MPA was ume injector), and using alternate detection formats found bound to the serine residue. In addition, the (eg, flame photometric detection, positive ion chemical procedure was useful for detecting BChE adducted ionization, and high-resolution electron impact MS).6,33 to OP pesticides as well as non-OP anti-ChEs, such Additionally, Jakubowski et al34 have successfully ap- as pyridostigmine.44 A limitation of the assay is that plied the procedure to RBCs. agent identity is needed for MS analysis.45 For this With regard to the ability of a fluoride ion to re- reason, an extension of this procedure was developed generate soman bound to BChE, it is well known that that uses a generic approach.45 The method employed the process of aging would preclude release from the a chemical modification of the phosphyl group on the enzyme. However, studies have indicated that the fluo- serine residue to a common nonapeptide, regardless ride ion regeneration process, as applied to soman-poi- of the specific agent involved.45 Because a common soned animals, has produced contrary results.35 These nonapeptide is the outcome, a single MS method was studies suggest that soman can be displaced from sites employed in the analysis (Table 22-3).45 where aging does not play a significant role. Black et al36 have demonstrated that both sarin and soman bind Application to Human Exposures to tyrosine residues of human serum albumin. The observation that the alkyl group remained intact, in The fluoride ion regeneration procedure29 was used particular for soman, argues that binding to this site to analyze serum from exposed individuals in the Aum does not result in aging as seen with ChEs.36 Similarly, Shinrikyo terrorist attacks at Matsumoto and in the carboxylesterase, known to exist in high quantities in Tokyo subway. As previously indicated, this procedure rats and mice, has been shown to form adducts with is based upon the use of fluoride ion to regenerate the soman.37–41 Moreover, soman formation has been parent agent and free BChE. The amount of regener- demonstrated via fluoride-induced regeneration of ated sarin from serum ranged from 1.8 to 2.7 ng/mL soman-inhibited carboxylesterase in rat plasma37 and in the Matsumoto incident and 0.2 to 4.1 ng/mL in the purified human albumin.35 Although the presence of Tokyo attacks.29 carboxylesterase in significant amounts is questionable Although unable to detect MPA or IMPA directly in humans, the albumin provides a potential source of from the blood of victims of the Tokyo subway attack, the protein from which the agent can be regenerated. Nagao et al42,43 detected these compounds after alkaline Currently the utility of fluoride regeneration in human phosphatase digestion of the sarin-AChE complex. exposures involving soman is unclear. More studies are However, the authors did not report MPA or IMPA needed to clarify the utility of fluoride regeneration in concentrations. Although not directly relevant to di- humans with soman exposure following confirmed agnostic testing, a conceptually similar approach was events. also applied to formalin-fixed brain tissues (GB-bound Nagao et al42,43 employed a different approach, AChE) from victims of the Tokyo subway attack.1 The exploiting sarin bound to AChE, using blood as the assays conducted on frozen cerebral cortex did not de- matrix. This procedure detected IMPA, following its tect MPA or IMPA. Similar studies with formalin-fixed release from the sarin-AChE complex, using an al- cerebellum tissue resulted in detecting only MPA. The

699 Medical Aspects of Chemical Warfare

Table 22-3 Analytical Methods Using Adducts to Biomolecules

Sample Matrix Product Identified Analytical Method

Plasma/serum GA, GB GC-NPD1,2 Red blood cell IMPA, MPA GC-MS3,4 Brain (cerebellum) MPA GC-MS5 Plasma/serum VX-G GC-FPD/GC-MS6 Plasma/serum Phosphylated nonapeptides from BChE LC-MS-MS7 Plasma/serum GA, GB, GF, VX-G GC-MS/GC-MS(HR)8 Plasma/serum/red blood cell GB GC-MS9 Plasma/serum Phosphylated nonapeptides from BChE- derivatized lc-MS-MS10

BChE: butyrylcholinesterase FPD: flame photometric detection GA: tabun GB: sarin GC: gas chromatography GF: cyclosarin HR: high resolution IMPA: isopropyl methylphosphonic acid LC: liquid chromatography MPA: methylphosphonic acid MS: mass spectrometry NPD: nitrogen-phosphorus detector VX-G : ethyl methylphosphonofluoridate Data sources: (1) Polhuijs M, Langenberg JP, Benschop HP. New method for retrospective detection of exposure to organophosphorus anticholinesterases: application to alleged sarin victims of Japanese terrorists. Toxicol Appl Pharmacol. 1997;146:156–161. (2) Polhuijs M, Langenberg JP, Noort D, Hulst AG, Benschop HP. Retrospective detection of exposure to organophosphates: analyses in blood of human beings and rhesus monkeys. In: Sohns T, Voicu VA, eds. NBC Risks: Current Capabilities and Future Perspectives for Protection. Dordrecht, Holland, Netherlands: Kluwer Academic Publishers; 1999:513–21. (3) Nagao M, Takatori T, Matsuda Y, et al. Detection of sarin hydrolysis products from sarin-like organophosphorus agent-exposed human erythrocytes. J Chromatogr B Biomed Sci Appl. 1997;701:9–17. (4) Nagao M, Takatori T, Matsuda Y, Nakajima M, Iwase H, Iwadate K. Definitive evidence for the acute sarin poisoning diagnosis in the Tokyo subway. Toxicol Appl Pharmacol. 1997;144:198–203. (5) Matsuda Y, Nagao M, Takatori T, et al. Detection of the sarin hydrolysis product in formalin-fixed brain tissues of victims of the Tokyo subway terrorist attack. Toxicol Appl Pharmacol. 1998;150:310–320. (6) Jakubowski EM, Heykamp LS, Durst HD, Thompson SA. Preliminary studies in the formation of ethyl methylphosphonofluoridate from rat and human serum exposed to VX and treated with fluoride ion. Anal Lett. 2001;34:727–737. (7) Fidder A, Hulst AG, Noort D, et al. Retrospective detection of exposure to organophosphorus anti-cholinesterases: mass spectrometric analysis of phosphylated human butyrylcholinesterase. Chem Res Toxicol. 2002;15:582–590. (8) Degenhardt CE, Pleijsier K, van der Schans MJ, et al. Improvements of the fluoride reactivation method for the verification of nerve agent exposure. J Anal Toxicol. 2004;28:364–371. (9) Jakubowski EM, McGuire JM, Evans RA, et al. Quantitation of fluoride ion released sarin in red blood cell samples by gas chromatography-chemical ionization mass spectrometry using isoptope dilution and large-volume injection. J Anal Toxicol. 2004;28:357–363. (10) Noort D, Fidder A, van der Schans MJ, Hulst AG. Verification of exposure to organophosphates: generic mass spectrometric method for detection of human butyrylcholinesterase adducts. Anal Chem. 2006;78:6640–6644. inability to detect IMPA was due to hydrolysis during advantage of these methods stems from the relatively the 2-year storage period. The inability to detect hy- rapid agent elimination and resultant limited oppor- drolysis products in the cerebral cortex as opposed to tunity to obtain specimens. The advantage of using the cerebellum was reportedly consistent with the rela- adducts formed with large-molecular–weight targets tive AChE activity detected in each tissue. The study (AChE or BChE) is a longer time frame (relative to authors state that this is the first verification of nerve that of hydrolysis products) to verify exposures. Some agent exposure using formalin-fixed brains.1 investigations have indicated that methods employing Due to the limited number of human exposures to BChE provide benefits over those with AChE because nerve agents, it is difficult to fully ascertain the advan- BChE is more abundant in blood.44 Assays involving tages and disadvantages of various definitive testing BChE digestion with subsequent assay of nonapeptide methodologies. Numerous assays to detect hydrolysis fragments facilitate identification of aged or nonaged products in blood or urine have been developed; some adduct at the phosphylated serine-198 residue; there- have been employed in exposure incidents. The dis- fore they are potentially useful in detecting agents such

700 Medical Diagnostics

as soman. Few methods have been published that use reduces AChE and BChE activity.3 Thus, it is crucial to the assay of adducts to biomolecules to verify chemical diagnose OP exposure or intoxication early, and blood warfare agent exposure in humans (Table 22-4). ChE activity (usually RBC-AChE) can be exploited as a tool for confirming exposure to these agents and com- Cholinesterase Analysis mencing antidotal (oxime) therapy.48,49 Because these ChE inhibitors comprise a group of structurally diverse OP chemical warfare agents are potent and ir- compounds with a wide range of relative specificities reversible inhibitors. Exposure results in excessive for RBC-AChE and plasma BChE, a complete profile accumulation of acetylcholine that hyperstimulates of inhibition is probably more accurately reflected if cholinergic tissues and organs and ultimately leads both ChEs are measured. to life-threatening cholinergic crises in humans.3 The Exposure to OP nerve agents or pesticides that mechanism of OP toxicity is the inhibition of AChE results in inhibition of less than about 20% AChE and BChE involved in the termination of neurotrans- or BChE (especially if clinical symptoms are absent) mission in cholinergic synapses and neuromuscular may not easily be detected because of considerable junctions of the central nervous system.46 Synaptic inter- and intraindividual variations in AChE and AChE is not amenable to direct measurement, but (especially) BChE activities.50 Moderate clinical symp- because of functional similarities between synaptic toms of poisoning will be apparent at 50% to 70% and erythrocyte AChE, the activity of AChE in whole AChE inhibition, with severe toxicity seen at greater blood can be used as a reliable surrogate biomarker than 90% inhibition.51 While general measurement of of central and peripheral nervous system activity.47 ChE activity in blood is not specific for exposure to Exposure to OP nerve agents, carbamates, pesticides, any OP nerve agent, carbamate, or pesticide, labora- anesthetics, and drugs (such as cocaine) selectively tory measurements by MS techniques can positively

Table 22-4 Methods Used to Confirm Human Exposures to Nerve Agent Adducts to Biomolecules

Agent/Incident Sample Matrix Product Identified Concentration Reported Analytical Method

GB, Matsumoto and Serum GB Matsumoto (1.8–2.7 ng/mL) GC-NPD1,2 Tokyo, Japan T tokyo (0.2–4.1 ng/mL) GB, Tokyo, Japan Red Blood Cell impa, MPA NR GC-MS3,4 GB, Tokyo, Japan Brain (cerebellum) MPA N nR GC-MS5 GB, Tokyo, Japan plasma/Serum phosphylated nona- 10–20 pmol inhibited BChE/ml lc-MS-MS6 (selected samples) peptides from BChE

BChE: butyrylcholinesterase GB: sarin GC: gas chromatography IMPA: isopropyl methylphosphonic acid LC: liquid chromatography MPA: methylphosphonic acid MS: mass spectrometric NPD: nitrogen-phosphorus detector NR: not reported Data sources: (1) Polhuijs M, Langenberg JP, Benschop HP. New method for retrospective detection of exposure to organophosphorus anticholinesterases: application to alleged sarin victims of Japanese terrorists. Toxicol Appl Pharmacol. 1997;146:156–161. (2) Polhuijs M, Langenberg JP, Noort D, Hulst AG, Benschop HP. Retrospective detection of exposure to organophosphates: analyses in blood of human beings and rhesus monkeys. In: Sohns T, Voicu VA, eds. NBC Risks: Current Capabilities and Future Perspectives for Protection. Dordrecht, Holland, the Netherlands: Kluwer Academic Publishers; 1999:513–521. (3) Nagao M, Takatori T, Matsuda Y, et al. Detection of sarin hydrolysis products from sarin-like organophosphorus agent-exposed human erythrocytes. J Chromatogr B Biomed Sci Appl. 1997;701:9–17. (4) Nagao M, Takatori T, Matsuda Y, Nakajima M, Iwase H, Iwadate K. Definitive evidence for the acute sarin poisoning diagnosis in the Tokyo subway. Toxicol Appl Pharmacol. 1997;144:198–203. (5) Matsuda Y, Nagao M, Takatori T, et al. Detection of the sarin hydrolysis product in formalin-fixed brain tissues of victims of the Tokyo subway terrorist attack. Toxicol Appl Pharmacol. 1998;150:310–320. (6) Fidder A, Hulst AG, Noort D, et al. Retrospective detection of exposure to organophosphorus anti-cholinesterases: mass spectrometric analysis of phosphylated human butyrylcholinesterase. Chem Res Toxicol. 2002;15:582–590.

701 Medical Aspects of Chemical Warfare identify many OPs by evaluating their leaving group dianion) coincides with the Soret band of hemoglobin, from fluoride-reactivated proteins.28 Thus, the deter- resulting in interference and reduced assay sensitivity. mination of an individual’s ChE status can be impor- By increasing the wavelength from 412 nm to 436 nm, tant prior to decisions regarding oxime therapy and hemoglobin interference can be reduced by 75% and confirmation of OP poisoning, particularly in the case assay sensitivity improved without significant sacrifice of long-lasting (eg, VX), rapidly aging (eg, soman) of the indicator (TNB-) absorption.58 Additionally, the nerve agents and the surprising persistence of soman molar extinction coefficient for TNB (13.6 × 103 M-1 cm-1) in blood and tissues.52 used in the original Ellman assay57 has been widely Although plasma BChE activity is measured in applied to calculate ChE activities, however, this value occupational and clinical toxicology laboratories, it may vary depending on temperature, wavelength, and should be noted that BChE exhibits different kinetic buffer conditions.59 Changes in these experimental pa- properties with nerve agents and pesticides than does rameters can alter the extinction coefficient, resulting RBC-AChE.53 In humans, pesticide toxicity is well in different ChE activities from various laboratories. documented,54 and completely different profiles of There are many published variations of the original RBC-AChE and plasma BChE activities in pesticide- cuvette-based Ellman assay, including the 96-well poisoned individuals have been reported.55 Because microtiter plate format.60 decreased serum BChE often precedes a decline in Walter Reed Army Institute of Research Whole RBC-AChE, an assay that measures both ChEs is more Blood Assay. An important variation of the Ell- valuable for detecting initial (and smaller) changes in man ChE assay is the Walter Reed Army Institute ChE levels that may signify exposure. of Research Whole Blood Assay,61 which uses 4,4’- Several sensitive and specific assays for measuring dithiopyridine instead of DTNB as a chromogenic AChE activity in blood have been developed for use indicator and three thiocholine substrates (acetyl-, in clinical and toxicology laboratories. For routine butyryl-, and propionyl-thiocholine). The absorp- use, however, a number of drawbacks are apparent, tion maxima in the ultraviolet range (324 nm) of the including time-consuming sample preparation and 4-thiopyridone formed yields a high signal-to-noise long turn-around times. There is also a lack of stan- ratio because hemoglobin interference is minimal.62 dardization because of the difficulty comparing results The use of three substrates in this assay rapidly and between laboratories that use different ChE assays and simultaneously provides redundancy and independent report values in different or nonstandard units. None measurement of the RBC-AChE activities and plasma of the widely used methods has been approved by the BChE in a small sample of unprocessed whole blood, US Food and Drug Administration (FDA). Clinical using a 96-well microtiter plate spectrophotometer. determination of AChE and BChE activities in blood The method is not labor intensive, and although it can commonly uses several techniques (colorimetric, elec- be performed manually, it has been semi-automated trometric, and radiometric) and normally measures using a Beckman-Coulter robotic platform for high either RBC-AChE or serum BChE concentrations, but sample throughput. A unique feature of the Walter usually not both.56 Reed Army Institute of Research method is that the blood is not treated prior to assay (Figure 22-3). Thus, Colorimetric ChE Assays in the Clinical Laboratory both AChE and BChE activities can be obtained without centrifugation or the use of inhibitors for whole, Several ChE assays are based on the enzyme-linked frozen, or lysed blood specimens.62 production of colored products. For these assays, moni- Test-Mate Assay. In addition to the laboratory-based toring color production as a function of time directly methods, a field-deployable test unit is commercially reflects enzyme activity or the ability of the enzyme available, the Test-Mate ChE system (EQM Research to turn over substrate. Decreased enzyme turnover is Inc, Cincinnati, OH). Further information (instructions, reflected in less color production per unit time. description, clinical trial date, etc) on the kit is avail- Ellman Assay. The Ellman method57 is a popular able through the manufacturer.63 The method is also colorimetric procedure for detecting and monitoring based on the Ellman procedure and is supplied as a pesticide exposure. The breakdown of thiocholine kit containing reagents to measure erythrocyte AChE substrates (acetylthiocholine and butyrylthiocholine) and plasma BChE separately using a battery-oper- by AChE and BChE is detected kinetically using the ated photometric analyzer.64 A specific serum BChE Ellman reagent DTNB (5,5’-dithio-bis-2-nitrobenzo- inhibitor (As1397, or 10-[α-diethylaminopropionyl]- ate). This assay is accurate, reliable, and inexpensive. phenothiazone) is included in the kit and is required However, the absorption maxima (412 nm) of the re- to measure AChE (after a period of incubation). Two sulting yellow TNB- (3-carboxy-4-nitrobenzenethiolate capillary tubes containing whole blood samples are

702 Medical Diagnostics

Human blood collection in heparin/EDTA (control or organophosphorus-exposed) for cholinesterase measurement

WRAIR whole blood assay1 US Army Delta pH, Michel assay2 Test-Mate ChE assay3

Blood volume finger prick (minumum Approx 1 mL of whole blood to 20 μL whole blood minimum volume 10 μL) for AChE and BChE centrifuge, only AChE routinely for AChE and BChE measured

Blood can be frozen or kept at 4°C; Blood suitable for analysis within Fresh blood for immediate analysis; hemolysis not problematic 15 days when kept at 4°C; blood kept at 4°C to minimize AChE no hemolysis tolerated reactivation can be used

No sample preparation required; Blood centrifuged at 1,000 g for Incubate whole blood with specific whole blood used 20 min at 4°C, plasma discarded and inhibitor for plasma BChE (As1397) packed RBCs retained for analysis for measurement of AChE

WRAIR assay using standard 96 well Completely manual (AChE values Manual assay kit using a hand-held format and UV plate reader, manual recorded by hand every 20 sec) photometric analyzer to obtain and semi-automated (Biomek robotic using pH meter at 25°C AChE and BChE values separately platform), determines both AChE and at ambient temperature BChE activity simultaneously at 25°C

Average sample throughput per Sample throughput: up to 51 samples Average sample throughput per hour: manual, 40; semiautomated every 38 min by staggering addition hour: 12. Each kit contains reagents/ robotic, 48; both AChE and BChE of substrate; AChE only tubes for 96 samples; AChE or BChE

Calculate activity in international Calculate activity in delta pH units Calculate activity in U/mL or U/g Hb enzyme units (U/mL) per h

Fig. 22-3: Laboratory and field cholinesterase assays routinely used by the US Army and their required blood processing. AChE: acetylcholinesterase As1397: 10-(α-diethylaminopropionyl)-phenothiazone BChE: butyrylcholinesterase ChE: cholinesterase EDTA: ethylenediaminetetraacetic acid g: gravity Hb: hemoglobin RBCs: red blood cells UV: ultraviolet Data sources: (1) Gordon RK, Haigh JR, Garcia GE, et al. Oral administration of pyridostigmine bromide and huperzine A protects human whole blood cholinesterases from ex vivo exposure to soman. Chemico Biol Interact. 2005;157–158:239–246. (2) Ellin RI, Burkhardt BH, Hart RD. A time-modified method for measuring red blood cell cholinesterase activity. Arch Environ Health. 1973;27:48–49. (3) Taylor PW, Lukey BJ, Clark CR, Lee RB, Roussel RR. Field verification of Test-mate ChE. Mil Med. 2003;168:314–319.

703 Medical Aspects of Chemical Warfare necessary for AChE and BChE determination (and peripheral tissue and RBC-AChE. However, PB does correction for hemoglobin content, which is also not penetrate the blood-brain barrier, and thus does measured in each blood sample; see Figure 22-3). The not afford protection against seizures and subsequent kit is easy to use by a relatively untrained operator neuropathological states induced by a nerve agent such and matches the sensitivity of the laboratory-based as soman.70 During the Gulf War (1990–1991), more Ellman methods, but has relatively low throughput than 100,000 US and allied troops received PB (a single because it is performed manually. Although the Test- oral dose of 30 mg given every 8 h) as a pretreatment Mate ChE kit is designed primarily for field use, where against exposure to soman. This method of pretreat- it is widely used for monitoring pesticide exposure ment was demonstrated62 in an FDA-supported clinical in agricultural workers, longer processing times are trial of human volunteers given PB as a single 30-mg required for complete AChE and BChE screening and dose (Table 22-5). Maximal RBC-AChE inhibition of hemoglobin measurement. about 27% was seen after 2.5 hours, with recovery of activity to almost 100% after 24 hours. To demonstrate Electrometric ChE Assay (delta pH) protection and sequestering of AChE, the volunteers’ PB-pretreated blood was exposed ex vivo to soman, A manual method still widely used by the US followed by PB and soman removal from the blood us- Department of Defense to measure RBC-AChE is ing a small spin column, and monitoring the recovery a modification65 of the end-point delta pH method of RBC-AChE (decarbamylation) (Figure 22-4). All of originally described by Michel.66 This assay moni- the AChE activity protected by PB pretreatment was tors the decrease in pH (using a simple pH electrode restored to control levels within 3 hours. Plasma BChE and pH meter) that occurs when AChE catalyzes the was also inhibited in the same volunteers, albeit to a hydrolysis of acetylcholine to choline and acetic acid. lesser extent (approximately 11%, about one third of The assay is initiated by the addition of substrate (ace- the inhibition observed for AChE).62 Although it was tylcholine) and the change in pH is monitored over 17 formally approved by the FDA as a specific pretreat- minutes. This method is slow and laborious, although ment for soman poisoning in February 2003,71 PB was throughput can be increased by staggering addition not ordered to be taken by troops during the 2003 of substrate to each sample (see Figure 22-3). Up to 51 Iraq war. blood samples can be analyzed in 38 minutes, and the Huperzine A: Potential Next Generation Peripher- RBC-AChE activity is reported as a change (delta) in ally and Centrally Acting Protection and Sequester- pH units per hour.67 However, the delta pH method ing of Cholinesterase. Another pretreatment drug, requires centrifugation of blood to pellet RBCs, fol- physostigmine, can cross the blood-brain barrier and lowed by removal of the plasma (containing BChE) protect brain ChE in addition to RBC-AChE. However, prior to analyzing AChE activity. Lysis of the blood psychological and behavioral side effects (although sample is precluded, so blood must be iced (not frozen) partially offset by a low dose of scopolamine to block before centrifugation and AChE analysis. Furthermore, muscarinic cholinergic receptors) preclude its use as because plasma BChE is not determined, the complete an effective pretreatment against OP exposure. Fur- spectrum of blood inhibition is unknown. Although thermore, its use as an Alzheimer’s drug to improve this technique was developed nearly 60 years ago, short-term memory has been discontinued because it is reliable, and the US Army (through the Depart- of multiple unwanted side effects, including nausea, ment of Defense Cholinesterase Reference Laboratory) dizziness, headaches, and sweating. has a quality assurance testing program for primary In contrast to PB, huperzine A (Hup A), an alkaloid RBC-AChE monitoring of more than 25,000 military isolated from the moss Huperzia serrata, is a reversible personnel per year.68 AChE inhibitor in both the peripheral and central nervous systems.72,73 Hup A has been shown to be supe- Pretreatment Therapy for Nerve Agent Poisoning: rior to physostigmine in its anti-ChE activity and has Protection and Sequestering of Cholinesterase a longer biological half-life in humans and animals. Hup A is currently undergoing extensive clinical trials Pyridostigmine Bromide. The US Army’s current in the United States as a drug for Alzheimer’s disease pretreatment against potential nerve agent poisoning (it is already used for this purpose in China), though is the reversible and fairly short-acting AChE inhibitor it is already sold as an over-the-counter nutraceutical pyridostigmine bromide (PB), which was reviewed supplement for memory enhancement (FDA approval earlier.69 PB is a quaternary ammonium compound that is not required for this type of sale). Pretreatment provides temporary protection (by carbamylation) of with Hup A (in addition to postexposure treatment

704 Medical Diagnostics

Table 22-5 Red Blood Cell and Acetylcholinesterase protection studies using pyridostigmine bromide and Huperzine A after ex-vivo exposure to soman

Procedure*† Techniques and Rationale

1) Obtain blood samples from human volunteers aliquot blood from subjects administered oral dose of 30 mg pyridostigmine bromide, 200 µg huperzine A, or no drug 2) Ex-vivo exposure to soman Incubated blood with 1 µM soman for 10 min at room temperature; incubate control samples (no soman) with saline

3) Remove free drug and soman Centrifuge samples through a C18 chromatography spin column to bind drug and remove from the blood 4) Allow time for AChE decarbamylation (PB) maintain postcolumn samples at room temperature for up to 24 hours or dissociation (Hup A) 5) Monitor time for regeneration of AChE after aliquot samples collected for AChE activity assay at indicated times decarbamylation of PB or reversible inhibition postcolumn by Hup A 6) Measure AChE activity Use WRAIR whole blood ChE assay to determine recovered enzyme activity

*Acetylcholinesterase and pyridostigmine bromide or huperzine A yield an acetylcholinesterase-drug complex in procedures 1 and 2. This reaction demonstrates sequestered enzyme, which temporarily inhibits the active site. †Over time, the acetylcholinesterase-drug complex becomes acetylcholinesterase plus pyridostigmine bromide (the decarbamylated form of acetylcholinesterase) or dissociated huperzine A in procedures 3 and 4. These reactions demonstrate the sequestered and then restored enzyme activity. AChE: acetylcholinesterase ChE: cholinesterase PB: pyridostigmine bromide Hup A: huperzine A

with atropine and an oxime) may represent a superior dissociates, which occurs after soman is cleared from treatment strategy for protection against chemical the blood. To illustrate this, Hup-A– inhibited blood warfare agent exposure and should be investigated (drawn 1.5 h after human volunteers were given a dose further. of 200 µg Hup A; Figure 22-5) was exposed ex vivo to Animal studies have been used to examine the ef- soman.62 Blood samples were then rapidly centrifuged ficacy of high doses of Hup A against OP nerve agent through a small column to remove any free soman and toxicity. A dose of Hup A (500 µg/kg) significantly Hup A, the latter binding to the column matrix while reduces soman lethality (protective ratios of 2–3 when allowing AChE and BChE to pass through the column. used alone) and inhibits blood and brain AChE by 60% Under these circumstances, any RBC-AChE not pro- to 70%.74 Unlike PB, Hup A can cross the blood-brain tected by Hup A would be irreversibly inhibited by barrier and in rats protects brain AChE from inhibi- soman. In contrast, the RBC-AChE protected by Hup tion by soman, thus preventing build-up of excessive A would dissociate over time, and the AChE activity acetylcholine leading to seizures and associated neu- would eventually be restored. In one study, after so- ropathological damage. In guinea pig hippocampus, man treatment, no AChE activity was observed by the natural Hup A in subchronic doses (yielding 20%–30% Walter Reed Army Institute of Research Whole Blood inhibition of RBC-AChE) has little or no affinity for Assay in either the placebo- or drug-treated volunteers muscarinic, nicotinic, or N-methyl-d-aspartate ex- (see Figure 22-5). After the spin column removal of citatory amino acid receptors, and does not induce free Hup A and soman, and a 4-hour period to allow neuropathological damage. for complete dissociation, the Hup-A–inhibited AChE Because Hup A reversibly binds to AChE, thus was restored to the level that was initially inhibited protecting the enzyme from reaction with OPs, the by the drug (about 60% inhibition by the drug before activity of the Hup-A–protected but inhibited AChE is the column compared to 54% returned AChE activity restored once the drug-AChE complex spontaneously after the column). AChE inhibition and sequestering

705 Medical Aspects of Chemical Warfare

Fig. 22-4. Effects of soman (GD) on acetylcholinesterase (placebo) (AChE) activity in whole blood from human volunteers 3.0 who had taken pyridostigmine bromide (PB; 30 mg tablet). Blood was drawn 2.5 hours after dosing, when red blood cell 29.9% Inhibition by PB (carbamylated AChE) (RBC) AChE is maximally inhibited by PB (blue bar, left of 2.5 dashed line). After GD exposure, PB and GD were removed 70.1% AChE activity remaining using a C18 chromatography spin column (postcolumn treatment). The PB-protected (carbamylated and sequestered) 2.0 AChE activity returned by 6 hours postcolumn. Green bars show the correlation between the initial percent inhibition of 33.7% AChE activity RBC-AChE by PB (solid red line) and the subsequent return 66.3% Inhibition 1.5 of AChE activity due to decarbamylation of the protected enzyme after removing PB and GD (green bars). While there was about 29.9% inhibition of RBC-AChE by PB before GD

AChE activity (U/mL) 1.0 exposure (blue bar with arrow), at 24 hours (the green bar with arrow) there was about 33.7% return in activity after the column, demonstrating protection of RBC-AChE by PB pretreatment. Error bars represent the mean plus or minus 0.5 the standard error of the mean. AChE: acetylcholinesterase GD: soman 0.0 3 h 6 h

PB: pyridostigmine bromide 24 h 0 min

RBC: red blood cell 30 min 60 min 24 h, no GD control, no GD GD (1 μ M, 10 min)

1. Precolumn 2. Postcolumn 1. treatment

by Hup A increased, compared to PB. Thus, Hup A is lowing the terrorist attacks in Japan in 1994 and 1995. highly effective in protecting RBC-AChE from ex-vivo Several severely poisoned victims (in cardiopulmo- soman exposure. nary arrest or in comas with generalized convulsions) As a pretreatment, Hup A is a specific and highly had plasma BChE activities of 20% of normal (80% selective RBC-AChE inhibitor, with serum BChE re- inhibited).77 However, information regarding the de- maining unaffected at physiological concentrations.75 layed or long-term subclinical effects of low-level or Thus, after exposure to an OP nerve agent, the effective trace amounts of sarin and other nerve agents, insec- bioscavenging capacity for serum BChE is preserved.76 ticides, and a variety of environmental chemicals (to This lack of BChE inhibition represents an additional which individuals may be asymptomatic) is relatively advantage of Hup A in preventing OP toxicity, and scarce, both in military and civilian environments. helps explain increased tolerance of Hup-A–pretreated Potential scenarios can be envisioned in which low animals to soman in comparison to animals pretreated or trace exposures become significant. Given the pos- with PB. sibility of urban terrorism involving chemical warfare Studies of toxicology and treatment for nerve agent OP agents, federal, state, and local authorities now exposure have predominantly focused on lethal and have a variety of sensitive and accurate ChE and OP supralethal doses of the agent. The acute and long-term detection assays to initiate appropriate containment, effects of sarin in humans were well documented fol- decontamination, and treatment measures.

SULFUR MUSTARD

Background ed after a suspected exposure were used for standard clinical assays and not preserved for later analysis. Analysis of specimens, such as blood or urine, fol- Highly sophisticated analytical methods required to lowing a suspected chemical warfare agent exposure verify chemical warfare agent exposure were not a has been rare. Historically, biomedical samples collect- part of the clinical laboratory methods inventory. More

706 Medical Diagnostics recently, a large portion of the biomedical samples ob- sulfur mustard or thiodiglycol (TDG), a hydrolysis tained from casualties of suspected exposure to sulfur product of sulfur mustard. Fortunately, some of the mustard have come from the Iran-Iraq war in the 1980s. blood components and urine specimens were frozen Many victims of that conflict who exhibited clinical and reanalyzed years later, after newer analytical meth- signs consistent with exposure to sulfur mustard ods were developed. Recently reported methods of were transported to hospitals in Europe for medical analysis generally indicate greater levels of sensitivity treatment. Prior to 1995 laboratory analysis of these than previous methods or target other biomarkers of specimens used methods to measure unmetabolized sulfur mustard exposure. In addition to samples from

(placebo, predose control) 2.5

2.0 54% AChE activity recovered 46% AChE Inhibition 60% Inhibition by Hup-A (reversibly inhibited AChE) 40% AChE activity remaining 1.5

placebo, no Hup-A 1.0 Hup-A treated, precolumn Hup-A postcolumn

0.5

0.0 0 1 h 2 h 3 h 4 h 0.5 h control, no GD GD (1 μ M, 10 min)

1. Precolumn 2. Postcolumn recovery (h) 1. treatment

Fig. 22-5. Effects of soman (GD) on acetylcholinesterase (AChE) activity in whole blood from human volunteers who were given an increasing dose of huperzine A (Hup A). Blood was drawn 1.5 hours after the final 200-μg dose (blue bar, left of dashed line). The red bars represent red blood cell AChE from volunteers prior to receiving a placebo, while the blue bars represent AChE from an individual receiving the 200-µg Hup A dose. In the first part, after soman treatment, no AChE activity is observed by the Walter Reed Army Institute of Research assay in either the placebo- or drug-treated volunteers (red bars, close to 0 U/mL). However, after the spin column removal of free Hup A and soman, and a 4-hour period to allow for complete dissociation, the Hup-A–inhibited AChE is restored to the level that was initially inhibited by the drug (green bar with arrow, about 60% inhibition by the drug before the column versus 54% returned AChE activity after the column). Error bars represent the mean plus or minus the standard error of the mean. Note the increased inhibition and sequestering of AChE by Hup A compared to pyridostigmine bromide (see Fig. 22-4). AChE: acetylcholinesterase Hup A: huperzine A

707 Medical Aspects of Chemical Warfare the Iran-Iraq war, a small number of samples from in- Analysis of Urine Samples dividuals exposed to sulfur mustard in laboratory and field situations were collected, stored, and analyzed There are currently five urinary metabolites of pri- using more recently developed methods. mary interest in verifying exposure to sulfur mustard. Some of sulfur mustard’s physical properties and Two of the metabolites, TDG and thiodiglycol sulfoxide biochemical reactions are addressed in this volume (TDG-sulfoxide), are derived from chemical hydroly- (see Chapter 8, Vesicants) and have been reviewed sis reactions (Figure 22-6). The other three products extensively elsewhere.78 Of primary importance in are formed following sulfur mustard’s reaction with the development of assays for sulfur mustard is the glutathione (GSH). Each of the five analytes has been formation of a highly reactive sulfonium ion that is identified in the urine of sulfur-mustard–exposed produced following cyclization of an ethylene group of individuals. sulfur mustard. The sulfonium ion readily reacts with nucleophiles, such as water, or combines with a variety Analytical Methods of nucleophilic sites in macromolecules. The resulting chemical reactions are able to produce a number of free Efforts to analyze specific biomarkers of sulfur mus- metabolites and stable adducts that can be exploited tard exposure in urine samples prior to 1995 targeted for analysis in blood, urine, and tissue samples.7,79,80 either unmetabolized sulfur mustard or TDG (Table This section focuses primarily on metabolites that 22-6). Vycudilik prepared urine samples by initially have been identified in biomedical samples from sulfur saturating them with sodium chloride followed by mustard casualties. organic extraction using diethylether. The organic por- Before the early 1990s analytical methods for veri- tion was evaporated under nitrogen and reconstituted fying exposure to sulfur mustard consisted of assays with methylene chloride. After the addition of silica for the unmetabolized compound or the hydrolysis gel, the methylene chloride was evaporated under product TDG. Since 1995 a number of significant nitrogen, reconstituted with solvent, and analyzed advances have occurred. Many new metabolites have using GC-MS.81 Vycudilik later modified the method been identified from specimens of sulfur-mustard–ex- to isolate possible conjugates of sulfur mustard.82 The posed individuals, and instrument advances, such primary difference in the latter study was the addition as the ability to interface LC with MS and the use of of strong acid to the urine samples. Urine samples tandem MS, have resulted in significant increases in were mixed with an equal amount of concentrated test sensitivity and selectivity. Most newer methods hydrochloric acid and saturated with sodium chloride. of verifying exposure to sulfur mustard require ex- Using steam distillation, the distillate was collected tensive sample processing prior to introduction into in ether. Following sodium chloride saturation of the analytical system. The use of MS has enabled the the aqueous layer, the ether layer was dried and the incorporation of isotopically labeled forms of analytes residue dissolved in methylene chloride and silica gel. for use as internal standards during the earliest stages Samples were analyzed using GC and high-resolution of sample preparation. This has resulted in greater MS. Vycudilik reported that the methods could not dis- reproducibility of assays and made them more ame- tinguish between sulfur mustard and its hydroxyethyl nable to quantitative analysis. Although the laboratory metabolites present in the urine samples.82 methods presented in this section are not considered Wils et al treated urine with concentrated hydro- routine or standard, the efforts by a small number of chloric acid to convert TDG back to sulfur mustard.83,84 laboratories worldwide that are active in this area of Two methods were reported, only the later method research have made the methods more attainable to a is described here. The urine was passed through two wider range of laboratories. C18 solid phase extraction cartridges. Next, a solution

CH2CH2Cl CH2CH2OH CH2CH2OH O CH2CH2OH

S S O S S CH CH Cl CH CH OH CH CH OH 2 2 2 2 2 2 O CH2CH2OH Sulfur Mustard (HD) Thiodiglycol (TDG) TDG sulfoxide TDG sulfone

Fig. 22-6. Hydrolysis of sulfur mustard to produce thiodiglycol, followed by oxidation reactions. HD: sulfur mustard TDG: thiodiglycol

708 Medical Diagnostics of deuterated TDG was passed through the same car- a control group of patients found levels of TDG at low tridges. The purified urine was mixed with concentrat- nanogram-to-milliliter concentrations. While most ed hydrogen chloride (HCl) and heated. Sulfur mus- of the control levels were approximately 5 ng/mL, tard was purged from the solution and trapped onto two individuals had levels that exceeded 20 ng/mL. a Tenax-TA adsorption tube. Analysis was performed These high background levels probably indicate that using a thermodesorption cold trap injector interfaced the method was also converting another analyte, in with GC-MS. Analysis of urine samples obtained from addition to TDG, into sulfur mustard (see below).

Table 22-6 reports published prior to 1995 showing laboratory analysis of human biomedical samples following suspected exposure to sulfur mustard

Patient Sample Information* Unmetabolized Sulfur Mustard Hydrolysis Product†

Urine samples from 2 Iran-Iraq War casualties patient 1: 1.0 ng/mL NM treated at Vienna hospital; collected 7 days after patient 2: 1.5 ng/mL incident (no date given)1 Urine samples from 5 Iranian casualties treated at NM P patient C1: 90 ng/mL Ghent hospital; collected Patient C2: 45 ng/mL 10 days after incident (March 9, 1984)2 Patient C3: 40 ng/mL Patient C4: 40 ng/mL Patient C5: 15 ng/mL Control samples: 3–55 ng/mL Urine samples from 5 Iranian casualties treated at NM P patient range: 3–140 ng/mL Utrecht hospital; collected 10 days after incident C control samples: 3–55 ng/mL (March 9, 1984)2 Hair samples from 2 Iranian casualties; collected 1 patient 1: 0.5–1.0 µg/gram NM day after incident (Feb 27, 1986)3 Patient 2: not detected Autopsy specimens (tissues and body fluids) urine: not detected; Fat, Skin, Brain, nm from Iranian casualty treated at Munich Kidney: 5–15 mg/kg; Muscle, Liver, hospital; collected 7 days after incident (1985)4 Spleen, Lung: 1–2 mg/kg Urine samples from 12 Iran-Iraq War casualties 1–30 ng/mL for 6 individuals; not nm (1986); no other details provided5 detected in 6 individuals Urine samples from 7 Iranian casualties treated at nm 5–6 days postexposure: Ghent hospital; collected 5–6 and 18–19 days range 7–336 ng/mL; after incident (Feb 12–13, 1986)6 18–19 days postexposure: range 3–7 ng/mL Urine samples from 3 Iranian casualties treated at NM P patient range: 4–8 ng/mL Ghent hospital; collected 18–19 days after C control samples: 1–21 ng/mL incident (Feb 12–13, 1986)6 Urine samples from 8 Iranian casualties treated at NM P patient range: 5–76 ng/mL Utrecht hospital; collected 8–9 days after C control samples: 1–21 ng/mL incident (Feb 12–13, 1986)6

*These are the known details of the incident and sample collection time after suspected exposure. †The hydrolysis product was thiodiglycol. NM: not measured Data sources: (1) Vycudilik W. Detection of mustard gas bis(2-chloroethyl)-sulfide in urine. Forensic Sci Int. 1985;28:131–136. (2) Wils ERJ, Hulst AG, de Jong AL, Verweij A, Boter HL. Analysis of thiodiglycol in urine of victims of an alleged attack with mustard gas. J Anal Toxicol. 1985;9:254–257. (3) United Nations Security Council. Report of the mission dispatched by the Secretary-General to investigate allegations of the use of chemical weapons in the conflict between the Islamic Republic of Iran and Iraq. New York, NY: UN; 1986. Report S/17911. (4) Drasch G, Kretschmer E, Kauert G, von Meyer L. Concentrations of mustard gas [bis(2-chloroethyl)sulfide] in the tissues of a victim of a vesicant exposure. J Forensic Sci. 1987;32:1788–1793. (5) Vycudilik W. Detection of bis(2-chloroethyl)-sulfide (Yperite) in urine by high resolution gas chromatography-mass spectrometry. Forensic Sci Int. 1987;35:67–71. (6) Wils ERJ, Hulst AG, van Laar J. Analysis of thiodiglycol in urine of victims of an alleged attack with mustard gas, part II. J Anal Toxicol. 1988;12:15–19.

709 Medical Aspects of Chemical Warfare

Drasch et al examined urine samples for unmetabo- converts all target analytes into the single analyte TDG lized sulfur mustard. Following organic extraction, for analysis. All of the methods have similar limits of thin-layer chromatography, and derivativation with detection, approximately 0.5 to 1 ng/mL. Although an gold, the extracts were analyzed using electrothermal assay has been developed to analyze TDG-sulfoxide atomic absorption spectroscopy.85 separately without a conversion to TDG, the method More recently, additional methods have been is complicated by the high polarity of the analyte.90 developed for trace level analysis of TDG and TDG- Consequently, the more common approach is to use sulfoxide in urine.86–89 There are a number of charac- the reducing agent titanium trichloride. teristics common to the methods (Table 22-7). They all Unfortunately, regardless of the analytical method use GC in association with some form of MS analysis, used, background levels have consistently been found use a derivatizing agent to make the analyte more in urine samples obtained from nonexposed individu- amenable to GC analysis and to increase sensitivity, als. In the most extensive study of background levels, and incorporate an isotopically labeled form of TDG urine samples from 105 individuals were examined as an internal standard. Most of the methods use a for sulfur mustard metabolites using an assay that solid phase extraction cartridge for sample prepara- incorporates both a deconjugation process and a re- tion. Some of the methods incubate the urine samples duction step (ie, the assay measures free and bound with glucuronidase with sulfatase activity to release forms of both TDG and of TDG-sulfoxide).89 Quantifi- any glucuronide-bound conjugates. Some of the meth- able background levels were observed in 82% of the ods use titanium trichloride in hydrochloric acid to samples. Nearly 60% of the samples had observed reduce TDG-sulfoxide to TDG. The strong acid also levels of TDG in the 0.5- to 2.0-ng/mL range, while hydrolyzes acid-labile esters of TDG and TDG-sulf- approximately 9% had levels in the 10- to 20-ng/mL oxide (Table 22-8). Ultimately, each of the methods range. When urine samples with higher background

Table 22-7 Analysis Methods for Urine Samples to Measure Thiodiglycol or Thiodiglycol and Thiodiglycol-Sulfoxide

Glucuronidase TiCl3 Internal Detection Instrumentation Derivativizing Agent SPE Cartridge Incubation Reduction Standard Limit

2 1 Negative ion chemical pentafluorobenzoyl florisil yes No H4-TDG ~ 1 ng/mL ionization GC-MS chloride 2 2 Electron impact GC-MS heptafluorobutyric none yes No H8-TDG ~ 1 ng/mL anhydride 2 3 Negative ion chemical pentafluorobenzoyl florisil no Yes H4-TDG < 1 ng/mL ionization GC-MS-MS chloride 13 4 Positive chemical ioni- heptafluorobutyric oasis HLB yes Yes C4-TDG 0.5 ng/mL zation GC-MS-MS anhydride

13 C4: isotopically labeled carbon 2 H4: isotopically labeled hydrogen or deuterium 2 H8: isotopically labeled hydrogen or deuterium GC: gas chromatography HLB: hydrophilic-lipophilic-balanced MS: mass spectrometry SPE: solid phase extraction TDG: thiodiglycol

TiCl3: titanium trichloride Data sources: (1) Black RM, Read RW. Detection of trace levels of thiodiglycol in blood, plasma and urine using gas chromatography-electron- capture negative-ion chemical ionisation mass spectrometry. J Chromatogr. 1988;449:261–270. (2) Jakubowski EM, Woodard CL, Mershon MM, Dolzine TW. Quantification of thiodiglycol in urine by electron ionization gas chromatography-mass spectrometry. J Chromatogr. 1990;528:184–190. (3) Black RM, Read RW. Improved methodology for the detection and quantitation of urinary metabolites of sulphur mustard using gas chromatography-tandem mass spectrometry. J Chromatogr B Biomed Appl. 1995;665:97–105. (4) Boyer AE, Ash D, Barr DB, et al. Quantitation of the sulfur mustard metabolites 1,1’-sulfonylbis[2-(methylthio)ethane] and thiodiglycol in urine using isotope-dilution gas chromatography-tandem mass spectrometry. J Anal Toxicol. 2004;28:327–332.

710 Medical Diagnostics

Table 22-8 Target Analytes for Analysis Methods Outlined in Table 22-7

TDG TDG-Sulfoxide

Glucuronide-bound Acid-labile Glucuronide-bound Acid-labile Free conjugates esters Free conjugates esters

Yes Yes No No No No1 Yes Yes No No No No2 Yes No Yes Yes No Yes3 Yes Yes Yes Yes Yes Yes4

TDG: thiodiglycol Data sources: (1) Black RM, Read RW. Detection of trace levels of thiodiglycol in blood, plasma and urine using gas chromatography-electron- capture negative-ion chemical ionisation mass spectrometry. J Chromatogr. 1988;449:261–270. (2) Jakubowski EM, Woodard CL, Mershon MM, Dolzine TW. Quantification of thiodiglycol in urine by electron ionization gas chromatography-mass spectrometry. J Chromatogr. 1990;528:184–190. (3) Black RM, Read RW. Improved methodology for the detection and quantitation of urinary metabolites of sulphur mustard using gas chromatography-tandem mass spectrometry. J Chromatogr B Biomed Appl. 1995;665:97–105. (4) Boyer AE, Ash D, Barr DB, et al. Quantitation of the sulfur mustard metabolites 1,1’-sulfonylbis[2-(methylthio)ethane] and thiodiglycol in urine using isotope-dilution gas chromatography-tandem mass spectrometry. J Anal Toxicol. 2004;28:327–332.

levels were reanalyzed without the reduction step, • 1,1’-sulfonylbis[2-(methylsulfinyl)ethane] the TDG level in all the samples was less than 2.5 (SBMSE). ng/mL. This indicates that the free and bound forms of the TDG-sulfoxide, rather than the free and bound MSMTESE and SBMSE can be reduced using ti- forms of TDG, are responsible for a larger portion of tanium chloride and analyzed by GC-MS-MS as a the observed background levels in the nonexposed hu- single analyte: 1,1’-sulfonylbis[2-(methylthio)ethane] man urine samples. These results are consistent with (SBMTE; Exhibit 22-1).94,95 Black et al reported a limit of those found in other, smaller studies of background detection of 0.1 ng/mL,94 while Young et al extended levels.86,87,90,91 Boyer et al discovered that storage condi- the lower limit of detection to 0.038 ng/mL.95 To date, tion of urine samples must also be considered when no background levels have been found in the urine of analyzing samples for TDG and TDG-sulfoxide.89 In a unexposed individuals, including studies where urine study of urine samples stored at – 20oC for an 8-month samples from over 100 individuals were analyzed period, they found that all free and conjugated TDG in using two different assay methods.89,95 Alternatively, the samples had oxidized to free and conjugated TDG- MSMTESE and SBMSE can be analyzed individually sulfoxide. Consequently, the use of a reducing agent without reducing the two analytes to single analyte us- was shown to be critical for the analysis of samples ing electrospray LC-MS-MS.96 Lower limits of detection that had been frozen for any length of time. were 0.1 to 0.5 ng/mL for each of the analytes. Black et al identified a series of metabolites formed The final urinary biomarker to be discussed is also from sulfur mustard’s reaction with GSH, a small- a reaction product of sulfur mustard with GSH: 1,1’- molecular–weight tripeptide that acts as a free radical sulfonylbis[2-S-(N-acetylcysteinyl)ethane] (see Figure scavenger (Figure 22-7).91–93 While a large number of 22-7e; Table 22-9). Using solid phase extraction for metabolites were identified in animal experiments, sample cleanup and analyte concentration, followed there are three verified reaction products in urine by analysis with negative ion electrospray LC-MS-MS, samples obtained from individuals exposed to sulfur Read and Black were able to achieve detection limits mustard. One set of reaction products is believed to of 0.5 to 1.0 ng/mL.97 result from metabolism of the sulfur-mustard–GSH Assays for several other potential urinary analytes conjugate by the β-lyase enzyme (see Figure 22-7d). have been developed, but these analytes have yet to Two β-lyase metabolites have been identified in the be confirmed in human-exposed samples. N7-(2-hy- urine from exposed individuals: droxyethylthioethyl) guanine is a breakdown product from alkylated DNA that has been observed in animal • 1-methylsulfinyl-2-[2-(methylthio)ethylsulfo- studies. Fidder et al developed both a GC-MS method nyl]ethane (MSMTESE) and that requires derivatization of the analyte and an

711 Medical Aspects of Chemical Warfare

a glutathione (tripeptide)

O O

H2N CH CH2 CH2 C NH CH C NH CH2 COOH

COOH CH2SH

glutamic acid cysteine glycine

glutamic acid

NH O

b CH2CH2Cl CH2CH2 S CH2CH C glycine

S + 2 glutathione S

CH2CH2Cl CH2CH2 S CH2CH C glycine

NH O

glutamic acid

NH2

c CH2CH2 S CH2CH COOH

(O)nS

CH2CH2 S CH2CH COOH

NH2

β-lyase

HN COOH3 d e CH2CH2 SCH3 CH2CH2 S CH2CH COOH O2S O2S CH2CH2 SOCH3 CH2CH2 S CH2CH COOH

HN COOH3

CH2CH2 SOCH3

O2S

CH2CH2 SOCH3

Fig. 22-7. Reaction pathway proposed by Black et al (1992). (a) Structure of glutathione. (b) Reaction of sulfur mustard and glutathione. (c) Intermediate product. (d) β-lyase metabolites 1-methylsulfinyl-2-[2-(methylthio)ethylsulfonyl]ethane (MSM- TESE) and 1,1’-sulfonylbis[2-(methylsulfinyl)ethane] (SBMSE). (e) Bis-mercapturic acid conjugate of mustard sulfone. Data source: Black RM, Brewster K, Clarke RJ, Hambrook JL, Harrison JM, Howells DJ. Biological fate of sulphur mustard, 1,1’-thiobis(2-chloroethane): isolation and identification of urinary metabolites following intraperitoneal administration to rat. Xenobiotica. 1992;22:405–418.

712 Medical Diagnostics

LC-MS-MS method that can analyze the compound several days after admission to the hospital. Analysis directly.98 Other possible urinary analytes are an imid- using the same method produced negative results for azole derivative formed from sulfur mustard’s reaction all samples. with protein histidine residues99 and sulfur mustard Vycudilik also analyzed urine samples obtained adducts to metallothionien.100 from 12 Iran-Iraq War casualties.82 The only clinical description of the patients was the observation that Application to Human Exposure they had severe skin lesions resulting from an alleged mustard gas attack. Urine samples from six of the Vycudilik analyzed urine samples from two casual- patients produced positive results for sulfur mustard. ties of the Iran-Iraq War who were brought to a hospital Concentrations found ranged from 1 to 30 ng/mL. The in Vienna, Austria, for treatment of suspected exposure method could not distinguish between sulfur mustard to sulfur mustard.81 The exposure was believed to or its hydroxyethyl metabolites present in the urine have occurred 1 week prior to their arrival in Vienna. samples. No clinical description of the patients’ injuries was Wils et al examined a large number of urine samples provided in the report. The concentration of sulfur for TDG concentrations. The samples were obtained mustard found in their urine samples using GC-MS from Iranian casualties of the Iran-Iraq War transported was approximately 1.0 ng/mL and 1.5 ng/mL. Addi- to western European hospitals in Ghent and Utrecht tional urine samples were obtained from the patients for treatment.83,84 The majority of the urine samples

EXHIBIT 22-1 SAMPLE PREPARATION METHODS FOR THE Gas Chromatographic/Mass Spectro- metric/Mass Spectrometric ANALYSIS OF THE SULFUR MUSTARD URINARY β-LYASE METABOLITES

Procedure of Black et al: • Add the internal standard deuterated 1,1’-sulfonylbis[2-(methylsulfinyl)ethane] to 1 mL of urine. • Add 0.4 mL of titanium trichloride. • Incubate sample at 40°C overnight (16 hours).

• Filter solution through a preconditioned C8 Bond Elut solid-phase extraction cartridge. • Wash cartridge with water followed by a methanol and water mixture. • Allow cartridge to dry. • Elute analytes with acetone. • Evaporate to dryness under nitrogen and dissolve in toluene. • Analyze using GC-MS-MS with ammonia chemical ionization.1 Procedure of Young et al: • Place 0.5 mL of urine into a 15 mL tube. • Add the internal standard 13C-1,1’-sulfonylbis[2-(methythio)ethane] to the urine. • Add 1 mL of titanium trichloride. • Incubate sample at 75°C for 1 hour. • Add 2 mL of 6N sodium hydroxide and mix. • Centrifuge samples for 5 minutes. • Pour supernatant into a Chem Elut column. • Elute analytes with 16 mL of dichloromethane and acetonitrile mixture. • Evaporate to dryness under nitrogen and dissolve in toluene. • Analyze using GC-MS-MS with isobutane chemical ionization.2 GC: gas chromatography MS: mass spectrometry Data sources: (1) Black RM, Clarke RJ, Read RW. Analysis of 1,1’-sulphonylbis[2-(methylsulphinyl)ethane] and 1-methylsulphinyl- 2-[2-(methylthio)ethylsulphonyl]ethane, metabolites of sulphur mustard, in urine using gas chromatography-mass spectrometry. J Chromatogr. 1991;558:405–414. (2) Young CL, Ash D, Driskell WJ, et al. A rapid, sensitive method for the quantitation of specific metabolites of sulfur mustard in human urine using isotope-dilution gas chromatography-tandem mass spectrometry. J Anal Toxicol. 2004;28:339–345.

713 Medical Aspects of Chemical Warfare

Table 22-9 Analytical Methods Used to Verify Exposure to Sulfur Mustard in Biomedical Samples

Sample Matrix Biomarker Sample Preparation Analytical Method LOD

Urine, blood, tdG Enzyme incubation, derivatization negative ion chemical ~ 1 ng/mL1 plasma ionization GC-MS Urine tdG Enzyme incubation, derivatization electron impact GC-MS ~ 1 ng/mL2 3 Urine tdG, TDG-sulfoxide tiCl3 reduction, derivatization negative ion chemical < 1 ng/mL ionization GC-MS-MS 4 Urine tdG, TDG-sulfoxide enzyme incubation, TiCl3 reduction, positive ion chemical 0.5 ng/mL derivatization ionization GC-MS Urine tdG-sulfoxide derivatization Negative ion chemical 2 ng/mL5 ionization GC-MS 6 Urine SBMTE TiCl3 reduction Positive ion chemical 0.1 ng/mL ionization GC-MS-MS 7 Urine SBMTE TiCl3 reduction Positive ion chemical 0.04 ng/mL ionization GC-MS-MS Urine mSMTESE SPE cartridge extraction Positive ion electrospray 0.1–0.5 ng/mL8 LC-MS-MS Urine SBMSE SPE cartridge extraction Positive ion electrospray 0.1–0.5 ng/mL8 LC-MS-MS Urine Bis-(N-acetyl cysteine) SPE cartridge extraction Negative ion electrospray 0.5–1 ng/mL9 conjugate LC-MS-MS Blood hemoglobin valine Globin isolation, valine cleavage by negative ion chemical 100 nM whole adduct Edman degradation, derivatization ionization GC-MS blood expo- sure10,11 Blood hemoglobin valine Globin isolation, valine cleavage by high-resolution negative 0.5 pmol adduct Edman degradation ion chemical ionization adduct/mL12 GC-MS Blood hemoglobin histidine acid hydrolysis of globin, positive ion electrospray not reported12 adduct derivatization LC-MS-MS Blood hemoglobin histidine acid hydrolysis of globin, positive ion electrospray 10 µM whole adduct derivatization LC-MS-MS blood expo- sure13 Plasma albumin cysteine albumin isolation, pronase positive ion electrospray 10 nM whole adduct digestion LC-MS-MS blood expo- sure14,15 Blood, plasma protein adducts protein precipitation, alkaline negative ion chemical 25 nM plasma hydrolysis, derivatization, SPE ionization GC-MS exposure16 extraction Blood dna adducts WBC isolation, lysis, extraction, immunuslotblot assay 50 nM whole treatment with RNase and blood expo- proteinase K sure17 Skin DNA adducts epidermal layer isolation, lysis, immunuslotblot assay 1 sec skin extraction, treatment with RNase exposure to and proteinase K saturated vapor17 Skin Keratin adducts alkaline hydrolysis, derivatization lc-radiometric detector not reported18

(Table 22-9 continues)

714 Medical Diagnostics

Table 22-9 continued

Cl: chemical ionization DNA: deoxyribonucleic acid GC: gas chromatography LC: liquid chromatography LOD: limit of detection MS: mass spectrometry MSMTESE: 1-methylsulfinyl-2-[2-(methylthio)ethylsulfonyl]ethane RNase: ribonuclease SBMSE: 1,1’-sulfonylbis[2-(methylsulfinyl)ethane] SBMTE: 1,1’-sulfonylbis[2-(methylthio)ethane] SPE: solid phase extraction TDG: thiodiglycol

TiCl3: titanium trichloride WBC: white blood cell Data sources: (1) Black RM, Read RW. Detection of trace levels of thiodiglycol in blood, plasma and urine using gas chromatography-electron- capture negative-ion chemical ionisation mass spectrometry. J Chromatogr. 1988;449:261–270. (2) Jakubowski EM, Woodard CL, Mershon MM, Dolzine TW. Quantification of thiodiglycol in urine by electron ionization gas chromatography-mass spectrometry. J Chromatogr. 1990;528:184–190. (3) Black RM, Read RW. Improved methodology for the detection and quantitation of urinary metabolites of sulphur mustard using gas chromatography-tandem mass spectrometry. J Chromatogr B Biomed Appl. 1995;665:97–105. (4) Boyer AE, Ash D, Barr DB, et al. Quantitation of the sulfur mustard metabolites 1,1’-sulfonylbis[2-(methylthio)ethane] and thiodiglycol in urine using isotope-dilution gas chromatography-tandem mass spectrometry. J Anal Toxicol. 2004;28:327–332. (5) Black RM, Read RW. Methods for the analysis of thiodiglycol sulphoxide, a metabolite of sulphur mustard, in urine using gas chromatography-mass spectrometry. J Chromatogr. 1991;558:393–404. (6) Black RM, Clarke RJ, Read RW. Analysis of 1,1’-sulphonylbis[2-(methylsulphinyl)ethane] and 1-methylsulphinyl-2-[2-(methylthio)ethylsulphonyl]ethane, metabolites of sulphur mustard, in urine using gas chromatography-mass spectrometry. J Chromatogr. 1991;558:405–414. (7) Young CL, Ash D, Driskell WJ, et al. A rapid, sensitive method for the quantitation of specific metabolites of sulfur mustard in human urine using isotope-dilution gas chromatography-tandem mass spectrometry. J Anal Toxicol. 2004;28:339–345. (8) Read RW, Black RM. Analysis of beta-lyase metabolites of sulfur mustard in urine by electrospray liquid chromatography-tandem mass spectrometry. J Anal Toxicol. 2004;28:346–351. (9) Read RW, Black RM. Analysis of the sulfur mustard metabolite 1,1’-sulfonylbis[2-S-(N-acetylcysteinyl)ethane] in urine by negative ion electrospray liquid chromatography-tandem mass spectrometry. J Anal Toxicol. 2004;28:352–356. (10) Fidder A, Noort D, de Jong AL, Trap HC, de Jong LPA, Benschop HP. Monitoring of in vitro and in vivo exposure to sulfur mustard by GC/MS determination of the N-terminal valine adduct in hemoglobin after a modified Edman degradation. Chem Res Toxicol. 1996;9:788–792. (11) Noort D, Fidder A, Benschop HP, de Jong LP, Smith JR. Procedure for monitoring exposure to sulfur mustard based on modified Edman degradation of globin. J Anal Toxicol. 2004;28:311–315. (12) Black RM, Clarke RJ, Harrison JM, Read RW. Biological fate of sulphur mustard: identification of valine and histidine adducts in haemoglobin from casualties of sulphur mustard poisoning. Xenobiotica. 1997;27:499–512. (13) Noort D, Hulst AG, Trap HC, de Jong LPA, Benschop HP. Synthesis and mass spectrometric identification of the major amino acid adducts formed between sulphur mustard and haemoglobin in human blood. Arch Toxicol. 1997;71:171–178. (14) Noort D, Hulst AG, de Jong LP, Benschop HP. Alkylation of human serum albumin by sulfur mustard in vitro and in vivo: mass spectrometric analysis of a cysteine adduct as a sensitive biomarker of exposure. Chem Res Toxicol. 1999;12:715–721. (15) Noort D, Fidder A, Hulst AG, Woolfitt AR, Ash D, Barr JR. Retrospective detection of exposure to sulfur mustard: improvements on an assay for liquid chromatography-tandem mass spectrometry analysis of albumin-sulfur mustard adducts. J Anal Toxicol. 2004;28:333–338. (16) Capacio BR, Smith JR, DeLion MT, et al. Monitoring sulfur mustard exposure by gas chromatography-mass spectrometry analysis of thiodiglycol cleaved from blood proteins. J Anal Toxicol. 2004;28:306–310. (17) Van der Schans GP, Mars-Groenendijk R, de Jong LP, Benschop HP, Noort D. Standard operating procedure for immunoslotblot assay for analysis of DNA/ sulfur mustard adducts in human blood and skin. J Anal Toxicol. 2004;28:316–319. (18) Noort D, Fidder A, Hulst AG, de Jong LP, Benschop HP. Diagnosis and dosimetry of exposure to sulfur mustard: development of a standard operating procedure for mass spectrometric analysis of haemoglobin adducts: exploratory research on albumin and keratin adducts. J Appl Toxicol. 2000;20(suppl 1):S187–S192.

were the first collected following hospital admission from a casualty who died 1 day after admission. Dur- 5 to 10 days after the suspected exposure. Willems ing the hospitalization, 18 to19 days after the exposure, detailed the patients’ medical histories.101 Briefly, the a second set of urine samples was collected from one injuries were described as moderate to severe and group. TDG levels in that set of samples were similar were consistent with sulfur mustard injury, includ- to observed background levels in control samples. ing erythema and fluid-filled vesicles. Urine samples TDG background levels were determined from urine were initially analyzed for intact sulfur mustard, but samples obtained from nonexposed individuals and were found to be negative. Following treatment of were generally less than 12 ng/mL, although two of the the urine samples with a strong acid to convert TDG control samples had levels of 21 ng/mL and 55 ng/mL. to sulfur mustard, the samples were analyzed using Elevated background levels in control samples may GC-MS. The TDG concentrations found in the first set indicate that this method also converts TDG-sulfoxide of samples collected at the hospital ranged between or some other analytes into sulfur mustard. 5 to 100 ng/mL for the majority of the samples. The Drasch et al examined urine samples obtained highest observed TDG concentration was 330 ng/mL during an autopsy of a sulfur mustard casualty for

715 Medical Aspects of Chemical Warfare unmetabolized sulfur mustard.85 The victim was an viduals and coded them C1 through C5.101 This group Iranian soldier, age 24, who died of complications from of individuals was reportedly exposed to exploding pneumonia 7 days after the suspected exposure. The bombs that generated black dust and rain. Decon- patient had been transferred to an intensive care unit tamination efforts consisted only of clothing removal, in Munich, Germany. Samples were taken during the although some victims may have also showered. Early autopsy and stored at – 20°C for 1 year prior to analy- symptomatology included eye and throat irritation sis. Despite very high concentrations of sulfur mustard along with respiration difficulties. Within 1 to 2 days, found in autopsy tissue specimens, sulfur mustard was victims developed erythema and small blisters on the not detected in the urine samples. skin, edema of the eyelids, photophobia, coughing, In 1990 Jakubowski et al received urine samples from dyspnea, and hemoptysis. The patients were admitted an accidental laboratory exposure to sulfur mustard.102 to a hospital 10 days after the suspected sulfur mustard A liquid flashpoint tester overheated, vaporized a exposure. Urine samples were obtained at that time. mixture that was thought to contain only a deconned Four of the five patients were discharged 26 to 37 solution, and exposed a chemist who had attempted to days after hospitalization. Patient C1 developed adult shut down the reaction. Nine hours after the incident, respiratory distress syndrome and was given ventila- the individual felt a burning sensation on his arms, tory support, but died in cardiovascular shock 15 days hands, neck, and face. Medical care was sought the after the original exposure event (5 days after hospital morning after blisters appeared on his hands and arms. admission). Wils et al coded the urine sample set from The erythematous and vesicated areas were estimated individuals C1 to C5 as G1 to G5 in their report and, to be less than 5% and 1% of the total body surface using the GC-MS assay described above, found con- area, respectively. The patient collected his total urine centrations of TDG at 90 ng/mL, 45 ng/mL, 40 ng/mL, output for a 2-week period. For the first 3 days, the 40 ng/mL, and 15 ng/mL for individuals 1 through patient’s total urine output was only about a third to 5, respectively.83 Using a different method measuring half that of the average adult daily output of 1.5 liters, TDG and TDG-sulfoxide as a single analyte followed but the patient had a normal output level over the next by GC-MS-MS analysis, Black and Read found TDG 10 days. The assay method used measured both free and TDG-sulfoxide levels of 69 ng/mL, 28 ng/mL, and and conjugated TDG.87 The maximum TDG urinary 33 ng/mL for individuals C1, C2, and C5, respectively. excretion rate was 20 µg/day on the third day. TDG C3 and C4 were not assayed because their samples concentrations of 10 ng/mL or greater were observed were insufficient. Control urine samples analyzed by in some samples for up to 1 week after the exposure. Black and Read produced a background level of 11 A rate constant was calculated for TDG concentration ng/mL.88 Urine samples from all five casualties were from days 4 through 10 and the half-life was found analyzed for β-lyase concentrations, with the highest to be 1.2 days. A great deal of intraday variability concentration found in patient C1. The concentrations was noted for the TDG urine concentrations. Conse- found in the urine from four of the casualties ranged quently, the collection of several urine samples per between 0.5 to 5 ng/mL, while the individual who died day is recommended. An attempt was also made to had a β-lyase concentration of 220 ng/mL.88 estimate the total amount of sulfur mustard on the Once again using the GC-MS-MS method that patient’s skin. The estimate was based on two assump- measures both TDG and TDG-sulfoxide as a single tions: 1) the assay for the free and conjugated TDG analyte, Black and Read analyzed urine samples from represents approximately 5% of the total amount of two casualties of an alleged sulfur mustard attack on sulfur-mustard–related products in the blood, and 2) the Kurdish town of Halabja in 1988.88 The patients the bioavailability factor from skin to blood is 10 (ie, had been transferred to London for medical treat- 10% of the sulfur mustard on the skin penetrated into ment and urine samples were collected 13 days after the blood). A total of 0.243 mg of TDG was recovered the alleged incident. Black and Read found combined over a 2-week period. This represents 4.86 mg in the TDG plus TDG-sulfoxide levels of 11 ng/mL for both blood, or 48.6 mg on the skin. patients, but also found similar concentration levels There are currently four instances of human expo- in control samples. Urine samples were also analyzed sure to sulfur mustard in which urine samples were for β-lyase concentrations using GC-MS-MS. Although subjected to several different assays in order to target the concentration of the β-lyase metabolites found in multiple urinary metabolites. The first report described both patients was near the limit of detection for the a small subset of urine samples previously analyzed assay, the analytes were clearly detectable and ranged by Wils et al83 and were later reanalyzed by Black and between 0.1 and 0.3 ng/mL.88 The urine samples were Read88 after storage at – 20°C for a 5-year period. Wil- later analyzed using the LC-MS-MS assay that can lems provided clinical information on the five indi- distinguish the individual β-lyase metabolites.96 The

716 Medical Diagnostics

monosulfoxide MSMTESE was only detected from one oily liquid that was found leaking from the remnants of the casualties and was near the limit of detection of the munition. During the disposal operation, none for the assay. The bissulfoxide SBMSE was detected of the ordnance team members complained of eye or in the urine from both casualties, but for each sample throat irritation or breathing difficulties. The clini- it was near the limit of detection of the assay (0.1–0.5 cal sequence of events for one of the individuals has ng/mL). previously been reported,103,104 but will be described Some of the most extensive testing of urine samples in brief here. Within 2 hours of the munition destruc- for sulfur-mustard–related metabolites involved two tion, one of the individuals (patient D1, a 35-year-old individuals who were accidentally exposed to a World male) noticed a tingling sensation on one arm and then War I munition containing sulfur mustard. The injuries showered. The next morning (approximately 14 hours were described as predominately cutaneous exposures, after the liquid contact), painful areas had developed with both individuals exhibiting extensive skin blister- on his hand, along with noticeable reddening and small ing. Urine samples were collected 2 to 3 days after the blisters. He went to a local emergency room where the individuals were exposed. Black and Read analyzed blisters were observed to grow and coalesce. He was the urine using three different methods to detect me- subsequently transferred to a regional burn center. tabolites of sulfur mustard hydrolysis.91 In addition, Blisters developed on the patient’s arm, hand, ankle, the urine samples were examined for products of a and foot. The erythema and blistered area on the pa- reaction between sulfur mustard and GSH. The first as- tient was estimated to be 6.5% of his body surface area say measured TDG (free and conjugated together) and (Figure 22-8). The patient never developed ocular or found concentrations of 2 ng/mL for each individual. respiratory complications; consequently it appeared The second assay targeted only free TDG-sulfoxide, that his injuries were only the result of a cutaneous and concentrations of 69 ng/mL and 45 ng/mL were liquid exposure. The second individual (patient D2) found for the two individuals. Concentrations of 77 had a small, single blister and was not hospitalized. ng/mL and 54 ng/mL were found using a GC-MS- Urine samples from patient D1 were collected on days MS assay that measures TDG and TDG-sulfoxide as a 2 through 11 and 29, 35, and 42 days after exposure single analyte. Control samples analyzed along with and from patient D2 on days 2, 4, and 7 after exposure. the patient samples for the second and third assays Reports containing preliminary analysis results mis- gave levels of 4 to 5 ng/mL, therefore the patient takenly indicated that the first samples collected were results were significantly higher than control values. 1 day post exposure.104,105 Urine from both individuals The β-lyase metabolites were measured using both the was analyzed for hydrolysis metabolites and GSH GC-MS-MS method91 and the LC-MS-MS method.96 reaction products.104,105 Hydrolysis metabolites were When the β-lyase metabolites were analyzed indi- determined using several different methods. Using a vidually by LC-MS-MS, their concentrations ranged from 15 to 17 ng/mL and from 30 to 34 ng/mL for the monosulfoxide and bissulfoxide, respectively. When analyzed as the single, reduced form of SBMTE using GC-MS-MS, observed concentrations were 42 ng/ mL and 56 ng/mL. Samples were also analyzed for a bis-(N-acetylcysteine) conjugate using LC-MS-MS.97 This biomarker is also a reaction product of sulfur mustard and GSH. Although the biomarker is a major metabolite in rats exposed to sulfur mustard, it had not been reported in urine samples from exposed humans. The metabolite was found in urine samples from both exposed individuals, but concentrations were near the lower limit of detection (0.5–1 ng/mL). The most recently reported exposure incident involved two explosive ordnance technicians who were part of a team tasked with destroying a suspected World War I 75-mm munition. The munition had been discovered in a clamshell driveway and was believed to have originated from material dredged from a seafloor dumping area. Following demolition proce- Fig. 22-8. Aspiration of blister fluid from accidental exposure dures, two individuals came into contact with a brown to sulfur mustard.

717 Medical Aspects of Chemical Warfare

GC-MS assay that targets free TDG and glucuronide- has been verified in human exposure cases with sensi- bound TDG,87 detectable levels of TDG were only tive and selective assays. To date, no known examples observed in patient D1’s urine sample from day 2. of background levels of these metabolites have been Using a modified approach of the previous assay, the found in the urine from unexposed individuals (see urine samples were incubated with concentrated HCl Table 22-6, Table 22-10, Table 22-11). overnight to release acid-labile esters rather than sub- mitted to an enzyme incubation step. TDG levels for Analysis of Blood Samples patient D1 ranged from 40 ng/mL on day 2 to 10 ng/ mL on day 6 after exposure using the modified assay.104 Whereas urinary metabolites undergo relatively TDG was not detected in urine samples beyond day 6 rapid elimination from the body, blood components for patient D1 and was not detected in any of the urine offer biomarkers with potential use in verifying expo- samples from patient D2. Urine samples were also sure to sulfur mustard long after the exposure incident. analyzed using a GC-MS-MS assay that measured both Three different approaches have been used for blood free and glucuronide-bound TDG.105 There were no biomarker analysis. An intact macromolecule, such as detectable levels in any of the urine samples from the protein or DNA, with the sulfur mustard adducts at- individual with the single blister (patient D2). Patient tached, can be analyzed. Currently, this approach has D1 had the highest observed TDG concentration at day only been demonstrated for hemoglobin using in-vitro 2 (24 ng/mL). TDG concentrations ranged from 6 to experiments. An alternate approach is to enzymatically 11 ng/mL over the next 5 days, decreased to a range digest the proteins to produce a smaller peptide with of 1 to 2 ng/mL for the next 4 days, and TDG was the sulfur mustard adduct still attached. Methods of undetected after day 11. The final method of analysis this type have been developed for both hemoglobin for hydrolysis metabolites was a GC-MS-MS assay that and albumin. A third approach is to cleave the sulfur targeted free and glucuronide-bound TDG, free and mustard adduct from the macromolecule and analyze glucuronide-bound TDG-sulfoxide, and acid-labile it in a fashion similar to that used for free metabolites esters of TDG and TDG-sulfoxide.105 The observed found in urine. The latter two approaches, described concentrations for patient D2 (2–4 ng/mL) fell within below, have both successfully verified human exposure the range of concentrations previously observed in to sulfur mustard. urine samples of unexposed individuals. The highest observed levels found in unexposed individuals with Analytical Methods this assay were approximately 20 ng/mL.89 While observed concentrations were much higher for patient Methods to measure sulfur mustard adducts to D1, only days 2, 5, and 6 produced concentrations DNA in white blood cells have been developed us- (50, 28, and 24 ng/mL, respectively) that were greater ing LC with fluorescence detection106 and using an than the highest observed background control levels. enzyme-linked immunosorbent assay (ELISA).107,108 β-lyase metabolites were measured as the single ana- The DNA adduct that appears most abundant results lyte SBMTE using GC-MS-MS.105 Levels for both pa- from attachment of sulfur mustard to the N7 position tients decreased dramatically by day 3 after exposure. of deoxyguaninosine (Figure 22-9a).109 The immuno- Patient D2’s urine SBMTE concentrations were 2.6 ng/ chemical method used monoclonal antibodies that mL, 0.8 ng/mL, and 0.08 ng/mL for samples taken 2, were raised against N7-(2-hydroxyethylthioethyl)- 4, and 7 days after exposure, respectively. Patient D1’s guanosine-5’-phosphate (Exhibit 22-2). concentrations decreased from 41 ng/mL at day 2 after Hemoglobin is an abundant, long-lived protein in exposure down to 7 ng/mL, 3.3 ng/mL, and 1.3 ng/mL human blood. Alkylation reactions between sulfur over the next 3 days. For days 6 to 11, concentrations mustard and hemoglobin have been shown to occur ranged between 0.07 and 0.02 ng/mL and SBMTE was with six histidine, three glutamic acid, and two valine not detected beyond day 11. Patient D1’s urine from amino acids of hemoglobin.110,111 Methods have been days 2 and 3 was also examined for the presence of the developed to analyze several of the adducts. While bis-(N-acetylcysteine) conjugate using LC-MS-MS. It the histidine adducts appear to be the most abundant was detected at a concentration of 3.1 ng/mL in the type, their analysis using MS techniques is problematic urine sample collected 2 days after exposure, but was and the method does not appear to be as sensitive as not detected in the day 3 sample. the method for analyzing the N-terminal valine ad- Currently the analytes of choice for assessing poten- ducts.112 Adducts to the N-terminal valine amino acids tial exposure to sulfur mustard in urine samples are the represent only a small fraction of the total alkylation of two β-lyase metabolites. The analytes can be measured the macromolecule, but their location on the periphery individually using LC-MS-MS or reduced to a single of the molecule allows them to be selectively cleaved analyte (SBMTE) and measured using GC-MS-MS. This using a modified Edman degradation. Following

718 Medical Diagnostics isolation of the globin from the RBCs, the globin is compound can then be analyzed using negative ion reacted with pentafluorophenyl isothiocyanate to form chemical ionization GC-MS (Exhibit 22-3).113,114 a thiohydantoin compound, which is further deriva- Human serum albumin was found to be alkylated tized before analysis (Figure 22-10). The derivatized by sulfur mustard at the cysteine-34 position. Fol-

Table 22-10 Published Reports (1995–2006) of Laboratory Analysis of Human Urine Samples for Hydrolysis Metabolites Following Suspected Exposure to Sulfur Mustard

Patient Sample Glucuronidase Glucuronidase Incubation † ‡ § Information* Incubation TiCl3 Reduction TDG-sulfoxide & TiCl3 Reduction

Iranian casualties, 3 of 5 NM P patient C1: 69 ng/mL NM N nm individuals, treated at P patient C2: 28 ng/mL Ghent hospital, P patient C5: 33 ng/mL collected 10 days after C control: 11 ng/mL incident (March 9, 1984)1 Kurdish casualties, 2 NM P patient L1: 11 ng/mL NM N nm individuals, treated at P patient L2: 11 ng/mL London hospital, C control: 11 ng/mL collected 13 days after incident (March 17, 1988)1 Accidental exposure to patient S1: 2 ng/ml patient S1: 77 ng/ml patient S1: 69 ng/ml nm WWI munition, 2 patient S2: 2 ng/ml patient S2: 54 ng/ml patient S2: 45 ng/mL individuals, collected C control: 4.5 ng/ml control: 5 ng/mL 2–3 days after incident (1992)2 Accidental laboratory maximum excretion nm NM NM exposure, 1 individual, rate: 20 µg/day on collected 2–14 days day 3; concentration after incident (1990)3 > 10 ng/mL for 1 week postexposure Accidental exposure to patient D1: 24, 9, 5, 14, NM N M P patient D1: 50, 17, 11, 28, 24, WWI munition, 2 11, 6, 2, 2, 1.5, 1.2 14, 4.5, 9, 5, 6 ng/mL for individuals, collected ng/mL for days 2 days 2 to 11 after exposure, 2–42 days after incident to 11 after exposure, respectively (July 19, 2004)4 respectively Patient D2: not NM N M P patient D2: 1.8, 3, 4.4 ng/mL detected days 2, 4, 7 for days 2, 4, 7, respectively

*These are the known details of the incident and sample collection time after suspected exposure. †Assay measures TDG (free plus glucuronide-bound). ‡Assay measures free TDG, free TDG-sulfoxide, and acid-labile esters of both. §Assay measures TDG (free plus bound), TDG-sulfoxide (free plus bound), and acid-labile esters of both. NM: not measured TDG: thiodiglycol

TiCl3: titanium trichloride WWI: World War I Data sources: (1) Black RM, Read RW. Improved methodology for the detection and quantitation of urinary metabolites of sulphur mustard using gas chromatography-tandem mass spectrometry. J Chromatogr B Biomed Appl. 1995;665:97–105. (2) Black RM, Read RW. Biological fate of sulphur mustard, 1,1’-thiobis(2-chloroethane): identification of beta-lyase metabolites and hydrolysis products in human urine. Xeno- biotica. 1995;25:167–173. (3) Jakubowski EM, Sidell FR, Evans RA, et al. Quantification of thiodiglycol in human urine after an accidental sulfur mustard exposure. Toxicol Methods. 2000;10:143–150. (4) Barr JR, Young CL, Woolfit AR, et al. Comprehensive quantitative tandem MS analysis of urinary metabolites and albumin adducts following an accidental human exposure to sulfur mustard. In: Proceedings of the 53rd Conference of the American Society of Mass Spectrometry. San Antonio, Tex: June 5–9, 2005.

719 Medical Aspects of Chemical Warfare

Table 22-11 Published Reports (1995–2006) of Laboratory Analysis of Human Urine Samples for Glutathione Reaction Products Following a Suspected Exposure to Sulfur Mustard

Bis-(N-acetylcysteine) Patient Sample Information* β-lyase Metabolites† β-lyase Metabolites‡ Conjugate§

Iranian casualties, 5 of 5 individuals. patient C1: 220 ng/mL NM N nm treated at Ghent hospital, collected patient C2: 0.5 ng/mL 10 days after incident (March 9, patient C3: 1 ng/mL 1984)1 Patient C4: 5 ng/mL Patient C5: 1 ng/mL Kurdish casualties, 2 individuals, patient L1: 0.1 ng/ml patient L1: MSMTESE = nm treated at London hospital, <0.1 ng/mL, SBMSE = collected 13 days after incident ~ 0.1 ng/mL (March 17, 1988)1,2 Patient L2: 0.3 ng/ml patient L2: MSMTESE = 0.1 ng/mL, SBMSE = ~ 0.2 ng/mL Accidental exposure to WWI munition, patient S1: 42 ng/ml patient S1: MSMTESE = patient S1: 1 ng/mL 2 individuals, collected 2–3 days 15 ng/mL, SBMSE = 30 ng/mL after incident (1992)2,3,4 Patient S2: 56 ng/mL patient S2: MSMTESE = patient S2: 1 ng/mL 17 ng/mL, SBMSE = 34 ng/mL Accidental exposure to WWI munition, patient D1: 41, 7, 3.3, 1.3 NM P patient D1: 3.1 ng/mL 2 individuals, collected 2–42 days ng/mL for days 2–5 after incident (July 19, 2004)5 after exposure, respectively; 0.07–0.02 ng/mL for days 6–11 after exposure Patient D2: 2.6, 0.8, 0.08 ng/mL for days 2, 4, 7, respectively

*This information includes known incident information and sample collection time after suspected exposure. †Using GC-MS-MS analysis ‡Using LC-MS-MS analysis §Using LC-MS-MS analysis GC: gas chromatography LC: liquid chromatography MS: mass spectrometry MSMTESE: 1-methylsulfinyl-2-[2-(methylthio)ethylsulfonyl]ethane NM: not measured SBMSE: 1,1’-sulfonylbis[2-(methylsulfinyl)ethane] Data sources: (1) Black RM, Read RW. Improved methodology for the detection and quantitation of urinary metabolites of sulphur mustard using gas chromatography-tandem mass spectrometry. J Chromatogr B Biomed Appl. 1995;665:97–105. (2) Read RW, Black RM. Analysis of beta-lyase metabolites of sulfur mustard in urine by electrospray liquid chromatography-tandem mass spectrometry. J Anal Toxicol. 2004;28:346–351. (3) Black RM, Read RW. Biological fate of sulphur mustard, 1,1’-thiobis(2-chloroethane): identification of beta-lyase metabolites and hydrolysis products in human urine. Xenobiotica. 1995;25:167–173. (4) Read RW, Black RM. Analysis of the sulfur mustard metabolite 1,1’-sulfonylbis[2-S-(N-acetylcysteinyl)ethane] in urine by negative ion electrospray liquid chromatography-tandem mass spectrometry. J Anal Toxicol. 2004;28:352–356. (5) Barr JR, Young CL, Woolfit AR, et al. Comprehensive quantitative tandem MS analysis of urinary metabolites and albumin adducts following an accidental human exposure to sulfur mustard. In: Proceedings of the 53rd Conference of the American Society of Mass Spectrometry. San Antonio, Tex: June 5–9, 2005. lowing isolation of the albumin from the blood, the was analyzed using LC-MS-MS.115 The lower limit of albumin can be reacted with pronase enzyme to digest detection for the assay (1 nM) was once again reported protein. One of the resulting peptide fragments is a as an equivalent exposure level, as determined using tripeptide of the sequence cysteine-proline-phenyl- in-vitro exposures to sulfur mustard in human whole alanine, which contains the sulfur-mustard–alkylated blood. Recently, a modification to the isolation of the cysteine-34. After solid phase extraction, the tripeptide albumin from blood was reported using affinity chro-

720 Medical Diagnostics

a O Upon release, the adduct (in the form of TDG) was CH CH SCH CH OH 2 2 2 2 derivatized and analyzed using negative-ion chemical ionization GC-MS (Figure 22-11; see Exhibit 22-3). The N HN lower limit of detection for the assay in plasma was 25 nM, as determined using in-vitro exposures of sulfur mustard in human plasma.118 N N The limits of detection or equivalent levels of expo- H2N sure to sulfur mustard reported for most of the blood assays are based on in-vitro exposures of whole blood b NH2 or plasma. Quantitation of patient samples report the amount of sulfur mustard adducts that are found in N the samples relative to the amount of adducts that N are found from in-vitro exposures of whole blood or plasma at various known concentrations of sulfur mustard. Choosing whole blood over plasma to N N generate the in-vitro standard curves produces very different results because sulfur mustard readily reacts CH2CH2SCH2CH2OH with hemoglobin. Additionally, the technique used c for generating in-vitro standards can have significant O CH2CH2 S effects. For example, approximately a 30% difference was observed for the generation of two in-vitro standard curves, depending on how the sulfur mustard N 119 HN was incubated in blood. Higher adduct levels were observed when the sulfur mustard was allowed to react with the blood for 2 hours at 37°C, as opposed N N to 4 hours at room temperature. H2N 2 Application to Human Exposure

d OCH2CH2SCH2CH2OH Blood samples following a suspected human exposure to sulfur mustard rarely become available for N laboratory analysis. Three of the five known reports N involve the analysis of samples that were taken from casualties of the Iran-Iraq War, frozen for several years, then reanalyzed to verify exposure as new methods N N H2N H were developed. The other two published reports are on the analysis of blood samples obtained from three Fig. 22-9. Deoxyribonucleic acid adducts resulting from individuals who were casualties of accidental expo- reaction with sulfur mustard. (a) N7-HETE-guanine. (b) sures to World War I munitions. N3-HETE-adenine. (c) Bis[2-(guanin-7-yl)ethyl]sulfide. (d) The blood from two Iranian casualties who were be- O6-HETE-guanine. lieved to have been exposed to sulfur mustard in 1988 was analyzed using both the ELISA method for DNA adducts and the GC-MS method for the analysis of matography rather than the precipitation procedure.116 the N-terminal valine of hemoglobin.120 Samples were The modified procedure significantly reduced the collected 22 and 26 days following the suspected expo- sample preparation time. sure. One of the casualties had skin injuries that were The final method for analyzing blood samples to consistent with an exposure to sulfur mustard, but be discussed targets blood proteins in a more general the second casualty had injuries that were described approach. It was previously shown that sulfur mustard as only “vaguely compatible” with sulfur mustard adducts of glutamic and aspartic acids to keratin could exposure. Both individuals had approximately the be cleaved using base.117 Using a similar approach, same level of hemoglobin valine adduct, equivalent to precipitated proteins from plasma, whole blood, or the amount observed from a 900-nM, in-vitro, sulfur RBCs were treated with base to liberate the sulfur mus- mustard exposure in whole blood. ELISA DNA adduct tard adduct, hydroxyethylthioethyl, from the protein. levels observed in the granulocytes were also similar

721 Medical Aspects of Chemical Warfare

EXHIBIT 22-2 SAMPLE PREPARATION PROCEDURE FOR SULFUR MUSTARD ADDUCTS TO deoxyribonucleic acid IN BLOOD

Procedure of van der Schans et al for immunuslotblot analysis: • Collect blood specimen in vacutube containing EDTA. • Isolate and denature DNA using following procedure: - transfer 0.3 mL of blood to Eppendorf tube. - add RBC lysis solution, mix, and centrifuge. - lyse pelleted WBCs with cell lysis solution containing proteinase K. - Shake solution for 1 hour. - treat solution with RNAse. - add protein precipitation solution, mix, and centrifuge. - transfer supernatant to tube containing isopropanol and centrifuge. - Wash pellet with 70% ethanol, centrifuge, and air dry. - dissolve pellet overnight in Tris buffer containing HCl and EDTA. - determine DNA concentration using UV-VIS spectrometer. - prepare DNA solution with formamide, formaldehyde, and Tris buffer. - Incubate solution at 52°C. - Cool rapidly on ice and store at – 20°C. • Immunuslotblot assay procedure: - dilute denatured DNA samples in PBS. - Spot DNA solution onto nitrocellulose filter. - Wash spotted sample with PBS and air-dry filter. - cross-link DNA to filter using UV-gene-cross-linker. - incubate filter with blocking solution. - Wash with PBS/Tween solution. - incubate filter overnight with monoclonal antibodies that recognize N7-(2-hydroxyethylthioethyl)-2’- deoxyguanosine at 4°C with continuous shaking. - Wash with PBS/Tween solution. - incubate filter with rabbit-anti-mouse-Ig-horseradish peroxidase antibody for 2 hours at room temperature with shaking. - Wash with PBS/Tween solution. - Blot dry filter and transfer into a luminometer cassette. - measure luminescence.

DNA: deoxyribonucleic acid EDTA: ethylenediaminetetraacetic acid HCl: hydrochloric acid Ig: immunoglobulin PBS: phosphate-buffered saline RBC: red blood cell RNAse: ribonuclease Tris: trishydroxymethylaminomethane UV-VIS: ultraviolet-visible spectrophotometry WBC: white blood cell Data source: (1) Van der Schans GP, Mars-Groenendijk R, de Jong LP, Benschop HP, Noort D. Standard operating procedure for immunoslotblot assay for analysis of DNA/sulfur mustard adducts in human blood and skin. J Anal Toxicol. 2004;28:316–319.

722 Medical Diagnostics

O NH2 for both individuals, 150 to 160 nM. The individual with the skin injuries consistent with sulfur mustard exposure had observed ELISA DNA adduct levels in globin C CH + S(CH2CH2Cl)2 the lymphocytes that were only about half that observed in the individual with injuries that were less CH(CH3)2 pronounced: 220 nM and 430 nM, respectively. Blood samples obtained in 1986 from a group of a Iranian casualties that were treated at a hospital in Ghent for injuries believed to have been caused by sulfur mustard were examined years later using MS

O HN CH2CH2SCH2CH2OH methods by Black et al for both valine and histidine adducts of hemoglobin.114 The four individuals had blood samples collected at 5 or 10 days following the globin C CH suspected exposure event. Levels of the valine adduct ranged between 0.3 and 0.8 ng/mL. Observed

CH(CH3)2 levels of the histidine adduct were greater than the amount of valine adduct and ranged between 0.7 and 2.5 ng/mL. Using the same methods, Black et al also b C6F5NCS examined blood from one of the two individuals who were accidentally exposed to a World War I sulfur mustard munition.114 Several urinary metabolites were CH2CH2SCH2CH2OH detected in specimens from this individual, indicating exposure to sulfur mustard (see above). In a blood N sample obtained 2 days after the exposure, the valine S CH(CH3)2 and histidine hemoglobin adduct levels were 0.3 ng/ + globin mL and 2.5 ng/mL, respectively.

N Blood samples obtained from nine Iranian casualties of sulfur mustard exposure were analyzed for the N-terminal valine adduct of hemoglobin using C6F5 O GC-MS121 and for the albumin cysteine adduct using LC-MS-MS.115 All nine individuals were hospitalized c HFBI and had skin changes consistent with sulfur mustard exposure. Several of the casualties also had respiratory CH2CH2SCH2CH2OC(O)C3F7 difficulties. Blood samples were collected between 8 and 9 days after the exposure incident. Exposure levels N of the patient blood samples were correlated with hu- CH(CH ) S 3 2 man whole blood that was exposed to sulfur mustard in vitro. Adduct levels for both the hemoglobin valine N adduct and for the albumin cysteine adduct were in very close agreement with each other. Observed

C6F5 O exposure levels were between 0.3 and 2 µm and 0.4 and 1.8 µm for the hemoglobin and albumin adducts, respectively. Fig. 22-10. Analytical method of Fidder et al (1996) for The final exposure incident to be discussed involved blood exposed to sulfur mustard. (a) Reaction of N-terminal the two individuals who were accidentally exposed valine of globin with sulfur mustard. (b) Modified Edman to a World War I munition in 2004. Details of the degradation of N-terminal valine-sulfur mustard adduct. (c) exposure were given earlier in the urine section. This Derivatization using HFBI. particular human exposure to sulfur mustard differed HFBI: heptafluorobutyryl imidazole Data source: Fidder A, Noort D, de Jong AL, Trap HC, de from nearly all other previously reported incidents in Jong LPA, Benschop HP. Monitoring of in vitro and in vivo several important aspects. The two individuals are exposure to sulfur mustard by GC/MS determination of the only the second and third casualties of sulfur mustard N-terminal valine adduct in hemoglobin after a modified exposure to have both urine and blood samples made Edman degradation. Chem Res Toxicol. 1996;9:788–792. available for laboratory testing. Generally, urine or

723 Medical Aspects of Chemical Warfare

EXHIBIT 22-3 SAMPLE PREPARATION METHODS FOR THE Gas Chromatographic-Mass Spectro- metric ANALYSIS OF SULFUR MUSTARD ADDUCTS TO BLOOD BIOMOLECULES

Procedure of Fidder et al for sulfur mustard adduct to N-terminal valine of hemoglobin*: • Isolate globin from blood using following procedure: - centrifuge blood and remove plasma layer. - Wash RBCs with saline solution. - lyse RBCs with water and place solution into ice water bath. - centrifuge and transfer supernatant into tube containing HCl/acetone. - Wash the precipitate with HCl/acetone, acetone, and ether. - dry precipitate. • Mix isolated globin from blood sample with internal standard (globin isolated from blood that was previously exposed to deuterated sulfur mustard). • Dissolve globin in formamide. • Add pyridine and pentafluorophenyl isothiocyanate to solution. • Incubate solution at 60°C for 2 hours. • Add toluene, mix, centrifuge, and freeze samples in liquid nitrogen.

• Remove toluene layer and wash with water, aqueous Na2CO3, and water.

• Dry toluene with MgSO4, evaporate to dryness, and dissolve in toluene. • Filter solution through a preconditioned Florisil solid phase extraction cartridge. • Wash cartridge with dichloromethane. • Elute with methanol/dichloromethane. • Evaporate to dryness and dissolve in toluene. • Add heptafluorobutyryl imidazole and heat solution.

• After cooling, wash with water, aqueous Na2CO3, and water.

• Dry toluene with MgSO4 and concentrate solution. • Analyze using GC-MS with methane negative ion chemical ionization.1 Procedure of Capacio et al for sulfur mustard adducts to aspartic and glutamic acid residues of blood proteins: • Precipitate blood proteins: - for plasma samples, use acetone. - for whole blood or RBCs, use HCl/acetone. • Centrifuge solution and discard supernatant. • Wash protein pellet with acetone and ether. • Centrifuge solution and discard supernatant. • Dry protein at room temperature. • Add dried protein to NaOH solution. • Heat solution at 70°C for 1.5 hours. • Neutralize solution with HCl and dry with sodium sulfate. • Add ethyl acetate, mix, and remove ethyl acetate layer. • Add internal standard (deuterated thiodiglycol) to ethyl acetate and dry with sodium sulfate. • Add pyridine and pentafluorobenzoyl chloride. • After 10 minutes, add water and sodium bicarbonate. • Remove ethyl acetate layer and dry with sodium sulfate. • Pass ethyl acetate through a preconditioned silica solid-phase–extraction cartridge and collect the filtered solution. • Pass additional ethyl acetate through cartridge, collect, and combine the two fractions. • Analyze using GC-MS with methane negative ion chemical ionization.2 (Exhibit 22-3 continues)

724 Medical Diagnostics

Exhibit 22-3 continued

*The lower limit of detection for the assay was determined using in vitro exposures of sulfur mustard in human whole blood and was determined to be equivalent to a 100 nM exposure level.1,3 Following the administration of a single dose of sulfur mustard to a mar- moset (4.1 mg/kg; intravenous), the valine adduct was still detected in blood taken 94 days later.4 Intact hemoglobin with the sulfur mustard adducts attached have been examined using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry, but to date the technique has only been utilized for in vitro experiments at relatively high concentrations of sulfur mustard.5 GC: gas chromatography HCl: hydrochloric acid MS: mass spectrometry RBC: red blood cell Data sources: (1) Fidder A, Noort D, de Jong AL, Trap HC, de Jong LPA, Benschop HP. Monitoring of in vitro and in vivo exposure to sulfur mustard by GC/MS determination of the N-terminal valine adduct in hemoglobin after a modified Edman degradation. Chem Res Toxicol. 1996;9:788–792. (2) Capacio BR, Smith JR, DeLion MT, et al. Monitoring sulfur mustard exposure by gas chromatography-mass spectrometry analysis of thiodiglycol cleaved from blood proteins. J Anal Toxicol. 2004;28:306–310. (3) Noort D, Fidder A, Benschop HP, de Jong LP, Smith JR. Procedure for monitoring exposure to sulfur mustard based on modified edman degradation of globin. J Anal Toxicol. 2004;28:311–315. (4) Noort D, Benschop HP, Black RM. Biomonitoring of exposure to chemical warfare agents: a review. Tox Appl Pharmacol. 2002;184:116–126. (5) Price EO, Smith JR, Clark CR, Schlager JJ, Shih ML. MALDI-ToF/MS as a diagnostic tool for the confirmation sulfur mustard exposure. J Appl Toxicol. 2000;20(suppl 1):S193–S197.

blood samples that are collected and made available sulfur mustard. Concentrations of the plasma protein for verifying sulfur mustard exposure are from a single adducts were 97 nM on day 2 and decreased by 76% (23 time point after the exposure. In this instance, the pa- nM) by day 42, based on in-vitro exposures to sulfur tient with more severe injuries (patient D1, see above) mustard in human plasma instead of whole blood.122 had blood and urine collected almost daily for the first The assay could not detect plasma protein adducts in 10 days after exposure and then again on days 29, 35, patient D2. The assay was modified slightly to lower and 42. The incident also provided an opportunity the reported lower limit of detection of 25 nM, but to examine blood and urine metabolite levels from the limited amount of plasma received did not permit individuals with very different levels of injury. Patient D1 had extensive vesication of the arm and leg, while patient D2 developed only a single small blister. (The urinary metabolite results were detailed earlier in this protein COOH + S(CH2CH2Cl)2 section.) As expected, the observed concentrations of both urine hydrolysis metabolites and GSH reaction a products were much greater in patient D1. Sulfur mustard metabolite concentrations in the blood were protein COOCH CH SCH CH OH also much greater in patient D1. Blood metabolites 2 2 2 2 were assayed using two different methods. The first assay targeted the sulfur mustard adduct to cysteine- b NaOH 34 of albumin using pronase digestion of the protein 116 followed by LC-MS-MS analysis. Based on in-vitro protein + S(CH2CH2OH)2 exposures to sulfur mustard in human whole blood,

concentrations of albumin adducts found in the plasma c 2 C6F5COCl of patient D1 were 350 nM on day 2 after the exposure and had decreased by 74% (90 nM) on day 42.105 The rate of decrease over that time was consistent with S(CH2CH2OCOC6F5)2 the reported half-life of human albumin of 21 days. Albumin adduct concentrations for patient D2 over the Fig. 22-11. Analytical approach of Capacio et al. (a) Reaction sample collection period of 2 to 7 days after exposure of protein carboxylic acid groups with sulfur mustard. (b) remained stable and ranged between 16 and 18 nM. Hydrolysis of the ester groups to release thiodiglycol. (c) Derivatization of thiodiglycol using pentafluorobenzoyl The second assay targeted plasma protein adducts by chloride. cleaving the adduct with base, followed by analyzing Data source: Capacio BR, Smith JR, DeLion MT, et al. Moni- the derivatized adduct using negative ion chemical toring sulfur mustard exposure by gas chromatography-mass 118 ionization GC-MS. This was the first reported use of spectrometry analysis of thiodiglycol cleaved from blood this assay in the verification of a human exposure to proteins. J Anal Toxicol. 2004;28:306–310.

725 Medical Aspects of Chemical Warfare reanalysis for patient D2 using the modified method. mustard estimated to be between 0.5 and 1.0 µg/g. Blood provides several options for assessing po- The second patient’s hair sample tested negative for tential exposure to sulfur mustard because a variety sulfur mustard.123 of different metabolites have been verified in human Using the urine testing method to measure TDG exposure cases. Most of the assays are sensitive and levels, Wils et al assayed pieces of skin that were taken selective, and the majority of the methods use gas or from sulfur mustard casualties treated in the Ghent LC combined with MS. Background levels have not hospital. They found concentrations of TDG in the been found in the blood from unexposed individu- range of 2 to 7 µg/g in the skin samples.84 Immuno- als. The time period between exposure and sample chemical detection assays for sulfur mustard adducts collection and the severity of the injury are probably in human skin have been developed for keratin124 and the most important factors to consider when selecting DNA.108 To date, the assays have not been used on the appropriate assay. Currently the most sensitive skin samples following a suspected human exposure assay targets the alkylated cysteine of albumin, but to sulfur mustard. alkylated hemoglobin should offer a biomarker of Fluid removed from blisters resulting from sulfur greater longevity (Table 22-12). mustard exposure has also been analyzed from two casualties. Jakubowski et al obtained a small amount Analysis of Other Biomedical Sample Types of blister fluid from an accidental laboratory exposure (Tissue, Hair, Skin, Blister Fluid) that was detailed in the urine section above. The blister fluid was injected directly into a GC-MS for analysis. Urine and blood have been the traditional biomedi- Neither unmetabolized sulfur mustard nor TDG were cal samples used to test for exposure to sulfur mustard; observed in the blister fluid, but a polymer of TDG there are few reports on the analysis of other types of appeared to be present.102 The most recently reported biomedical samples. Using the same method described exposure also generated blister fluid samples that in the urine methods section, Drasch et al analyzed were made available for analysis (see Figure 22-8). several tissue types obtained from an autopsy for un- Korte et al examined blister fluid that was obtained metabolized sulfur mustard.85 An Iranian soldier, age on days 2 and 7 after exposure for both free TDG and 24, died of complications of pneumonia 7 days after a protein-bound adducts of sulfur mustard.122 Free TDG suspected exposure to sulfur mustard. The patient had concentrations were 19 ng/mL and 24 ng/mL for days been transferred to an intensive care unit in Munich, 2 and 7, respectively. The protein-bound adducts were Germany for treatment. Tissue samples were taken measured using the same method used for plasma during the autopsy, stored at – 20°C for 1 year, and then protein adducts.118 The observed levels were 63 and 73 analyzed for unmetabolized sulfur mustard. Sulfur pg/mg of protein for days 2 and 7, respectively. Blister mustard was found in the highest concentrations in fluid from day 7 was also assayed with the albumin the victim’s fat, skin, brain, and kidney; concentra- tripeptide LC-MS-MS method116 previously used for tions ranged from 5 to 15 mg/kg. Lesser amounts were plasma samples from the same individual. The concen- found in the patient’s muscle, liver, spleen, and lung tration of the alkylated tripeptide reported using this and ranged from approximately 1 to 2 mg/kg.85 assay was similar to the plasma concentration found Hair specimens have been analyzed for unme- for the same day.105 tabolized sulfur mustard using a methylene chloride Since 1995 there has been a significant increase extraction followed by analysis with GC-MS.123 Hair in the reported number of laboratory methods for samples that were obtained from two casualties of the verifying human exposure to sulfur mustard. Sen- Iran-Iraq War in 1986 were analyzed for unmetabolized sitivity to the analytical methods continues to im- sulfur mustard.123 The two casualties were examined prove and a number of new biomarkers have been by a United Nations inspection team as part of an identified and verified in biomedical samples from “alleged use” investigation initiated at Iran’s request. exposed individuals. These advances have helped The clinical history of the patients was consistent verify even low doses of sulfur mustard exposure with sulfur mustard exposure. Additionally, analysis and have extended the time period from exposure of vapor and soil samples from the bomb site tested event to collection. The major drawback to the current positive for sulfur mustard. The inspection team was laboratory methodologies is that they require expen- presented with hair samples from two individuals sive instrumentation, highly trained personnel, and that were reportedly exposed the previous day. The analytical standards that are generally not commer- hair specimens were sent to the National Defence cially available. Consequently, the time interval from Research Institute of Sweden for analysis. One of the sample collection, transport to an appropriate labora- hair samples produced a positive response for sulfur tory, sample preparation, instrument configuration,

726 Medical Diagnostics

Table 22-12 Published Reports (1997–2006) of Laboratory Analysis of Human Blood Samples Fol- lowing Suspected Exposure to Sulfur Mustard

Patient Sample Hemoglobin Hemoglobin Albumin Blood Protein Information* DNA Adduct† Adduct‡ Adduct§ Adduct¥ Adducts¶

Iranian casualties, 2 patient 1: Patient 1: 900 nM NM NM NM nm individuals, collected lymphocytes = 220 nM, 22 days (Patient 1) & granulocytes = 160 nM 26 days (Patient 2) after patient 2: Patient 2: 900 nM incident (1988)1 lymphocytes = 430 nM, granulocytes = 150 nM Iranian casualties, 4 nm Range: Range: NM N nm individuals, treated at 0.3–0.8 ng/mL 0.7–2.5 ng/mL Ghent hospital; collected 5 & 10 days after incident (1986)2 Accidental exposure to nm 0.3 ng/mL 2.5 ng/mL NM N nm WWI munition; 1 individual, collected 2 days after incident (1992)2 Iranian casualties, 9 nm Range: 0.3–2 μm nm Range: nm individuals, treated at 0.4–1.8 µM Utrecht hospital; collected 8–9 days after incident (1986)3,4 Accidental exposure to NM N M N nm patient D1: 350 patient D1: 97 WWI munition; 2 and 90 nM for and 23 nM for individuals, collected days 2 & 42 days 2 & 42 2–42 days after incident after exposure after exposure, (2004)5,6 respectively; respectively Patient D2: 16–18 nM for days 2, 4, 7 after exposure

*This information includes known incident information and sample collection time after suspected exposure. †N7-(2-HETE)-2’-deoxyguanosine ‡(HETE)-N-terminal valine § (HETE)-histidine ¥S-[2-(HETE)]-Cys-Pro-Phe ¶(HETE)-aspartic & glutamic acids DNA: deoxyribonucleic acid HETE: hydroxyethylthioethyl NM: not measured Data sources: (1) Benschop HP, van der Schans GP, Noort D, Fidder A, Mars-Groenendijk RH, de Jong LP. Verification of exposure to sulfur mustard in two casualties of the Iran-Iraq conflict. J Anal Toxicol. 1997;21:249–251. (2) Black RM, Clarke RJ, Harrison JM, Read RW. Biologi- cal fate of sulphur mustard: identification of valine and histidine adducts in haemoglobin from casualties of sulphur mustard poisoning. Xenobiotica. 1997;27:499–512. (3) Noort D, Hulst AG, de Jong LP, Benschop HP. Alkylation of human serum albumin by sulfur mustard in vitro and in vivo: mass spectrometric analysis of a cysteine adduct as a sensitive biomarker of exposure. Chem Res Toxicol. 1999;12:715–721. (4) Benschop HP, Noort D, van der Schans GP, de Jong LP. Diagnosis and dosimetry of exposure to sulfur mustard: development of standard operating procedures; further exploratory research on protein adducts. Rijswijk, The Netherlands: TNO Prins Maurits Laboratory. Final report DAMD17-97-2-7002, ADA381035, 2000. (5) Barr JR, Young CL, Woolfit AR, et al. Comprehensive quantitative tandem MS analysis of urinary metabolites and albumin adducts following an accidental human exposure to sulfur mustard. In: Proceedings of the 53rd Conference of the American Society of Mass Spectrometry. San Antonio, Tex: June 5–9, 2005. (6) Korte WD, Walker EM, Smith JR, et al. The determination of sulfur mustard exposure by analysis of blood protein adducts. Wehrmed Mschr. 2005;49:327.

727 Medical Aspects of Chemical Warfare and finally an analytical result is long. Because ex- developing diagnostic methods for the field and posure to sulfur mustard can occur without overt or patient care areas. Ongoing research in the use of immediate clinical signs and the onset of symptoms immunochemical assays is one area in particular that can be delayed for many hours, there is interest in offers hope for a field-forward assay.

LEWISITE

Background analysis.128 It is the most sensitive method reported to date, with a lower limit of detection of 7.4 pg/mL. Lewisite is an arsenical vesicant with a small Animal experiments have been limited in number molecular weight primarily found in the trans iso- and scope. In one study of four animals, guinea meric form, although it also exists in cis and geminal pigs were given a subcutaneous dose of lewisite (0.5 forms.125 Stockpiles of lewisite or lewisite mixed with mg/kg). Urine samples were analyzed for CVAA sulfur mustard reportedly exist in a number of coun- using both GC-MS and GC coupled with an atomic tries and present a potential risk for accidental expo- emission spectrometer set for elemental arsenic.129 sure. Most analytical methods that have been reported The excretion profile indicated a very rapid elimina- in the public scientific literature regarding lewisite or tion of CVAA in the urine. The mean concentrations related compounds are for the sample preparation detected were 3.5 μg/mL, 250 ng/mL, and 50 ng/mL and analysis of environmental samples. In the past for the 0- to 8-hour, 8- to 16-hour, and 16- to 24-hour 10 years, there have only been a handful of reports samples, respectively. Trace level concentrations regarding the analysis of biomedical samples to con- (0–10 ng/mL) of CVAA were detected in the urine firm lewisite exposure. Much like the other chemical of the 24- to 32-hour and 32- to 40-hour samples. warfare agents, lewisite is readily hydrolyzed in aque- The second animal study also used a subcutaneous ous solutions, including biological fluids. Therefore, dose of lewisite (0.25 mg/kg) given to four guinea the likelihood of finding the parent compound in a pigs.130 Using GC-MS, CVAA was detected in urine biomedical sample, such as blood or urine, is minimal. samples up to 12 hours following exposure. In the Consequently, method development has focused on same experiment, blood from the animals was also the breakdown compounds of lewisite or on prod- analyzed using GC-MS to detect CVAA. The amount ucts formed from its interaction with biomolecules. of measured CVAA was the sum of CVAA that was Until recently, most assays for lewisite have involved displaced from hemoglobin along with free CVAA in analyzing elemental arsenic using techniques such as the blood. The assay was able to detect the analyte atomic absorption spectroscopy. A drawback of this at 10 days after the exposure, although the concen- approach is that it lacks specificity because arsenic is tration was only 10% of that found 24 hours after ubiquitous in the environment. In addition to natu- exposure. Following the incubation of human blood rally occurring sources, arsenic is also found in some with radiolabeled lewisite, Fidder et al found that commercial products and food items (particularly 90% of the radioactivity was associated with the marine organisms). Arsenic is also a by-product of RBCs, and 25% to 50% was found with the globin.130 several industrial processes. Because of CVAA’s reactive nature, derivatization using a thiol compound has generally been applied Analysis of Urine and Blood Samples as part of the sample preparation process (Figure 22-12, Exhibit 22-4). Lewisite rapidly reacts with water to form chlorovinylarsonous acid (CVAA). CVAA slowly converts to Application to Human Exposure the arsinoxide form and polymerization reactions can occur. Earlier studies indicated that animals (species There are currently no reports of the collection of not specified) exposed to lewisite either topically or biomedical samples from individuals with suspected via injection were found to have measurable levels of lewisite exposure. Samples from such incidents are CVAA in their urine and throughout their digestive critical to confirm the validity of assaying for the systems.126 biomarkers observed in animal models. Additionally, Specific biomarkers of lewisite exposure are cur- the biomarkers that have been investigated in animal rently based on a limited number of in-vitro127,128 studies have indicated a rapid clearance of those and animal studies.129,130 Wooten et al developed a biomarkers in urine and less so for blood.130 This cre- solid phase microextraction headspace sampling ates problems for the retrospective determination of method for urine samples, followed by GC-MS lewisite exposure beyond a few days when analyzing

728 Medical Diagnostics

H As

C C Cl

Cl H

a H2O OH

H As

C C OH

Cl H

b EDT PDT

BAL

S CH2 S CH2

H As H As C H2 C C S CH2 C C S CH2

Cl H Cl H Logan (1999)1 Wooten (2002)2

S CH2

H As

C C S CH

Cl H CH2OH

c HFBI

S CH2

As 3 H Fiddler (2000) C C S CH

Cl H CH2OC(O)C3F7

Fig. 22-12. Published analytical approaches for the analysis of chlorovinylarsonous acid in urine. (a) Reaction of lewisite (trans isomer shown) with water to form chlorovinylarsonous acid. (b) Reactions of chlorovinylarsonous acid with thiol compounds ethanedithiol, propanedithiol, and British anti-Lewisite.1,2,3 (c) Derivatization using HFBI.3 BAL: British anti-Lewisite EDT: ethanedithiol

H2O: dihydrogen monoxide; water HFBI: heptafluorobutyryl imidazole PDT: propanedithiol Data sources: (1) Logan TP, Smith JR, Jakubowski EM, Nielson RE. Verification of lewisite exposure by the analysis of 2- chlorovinyl arsonous acid in urine. Toxicol Meth. 1999;9:275–284. (2) Wooten JV, Ashley DL, Calafat AM. Quantitation of 2- chlorovinylarsonous acid in human urine by automated solid-phase microextraction-gas chromatography-mass spectrometry. J Chromatogr B Analyt Technol Biomed Life Sci. 2002;772:147–153. (3) Fidder A, Noort D, Hulst AG, de Jong LP, Benschop HP. Biomonitoring of exposure to lewisite based on adducts to haemoglobin. Arch Toxicol. 2000;74:207–214.

729 Medical Aspects of Chemical Warfare

EXHIBIT 22-4 SAMPLE PREPARATION METHODS FOR Gas Chromatographic/Mass Spectrometric ANALYSIS OF chlorovinylarsonous acid

SPE procedure of Logan et al for CVAA in urine: • Precondition a C18 SPE cartridge with methanol followed by water. • Adjust pH of 1 mL of urine to pH 6 using HCl. • Add PAO as an internal standard. • Transfer sample to C18 SPE cartridge for sample cleanup/extraction. • Elute analytes using methanol. • Evaporate to dryness under nitrogen. • Reconstitute with ethanol containing EDT. • Analyze using GC-MS with EI ionization.1

SPME headspace procedure of Wooten et al for CVAA in urine: • Place 1 mL of urine into a 10-mL vial. • Add ammonium acetate buffer-water solution. • Add PAO as an internal standard. • Add PDT in order to derivatize CVAA and PAO. • Crimp-seal vial and heat at 70°C for 20 min. • Insert a 100-μM poly(dimethylsiloxane) SPME fiber into the sample vial headspace for 10 min. • Remove the SPME fiber from the vial and insert into the injection port of a GC-MS with EI ionization.2

SPE procedure of Fidder et al for free and bound CVAA in blood: • Add 2,3-dimercapto-1-propanol (BAL) to a 2-mL blood sample. The BAL causes displacement of bound CVAA and derivatizes both the free and released CVAA. • Add phenylarsine-BAL as an internal standard. • Shake the sample at room temperature overnight. • Dilute the sample with water. • Transfer sample to C18 SPE cartridge for sample cleanup/extraction. • Elute the analytes from the cartridge using dichloromethane/acetonitrile mixture. • Concentrate the extracted sample to dryness. • Redissolve in toluene. • Add HFBI to derivatize the BAL hydroxyl group. • Incubate the sample for 1 hour at 50°C and then cool to room temperature.

• Wash with water and then dry over MgSO4. • Analyze using GC-MS with EI ionization.3

BAL: British anti-Lewisite CVAA: chlorovinylarsonous acid EDT: ethanedithiol EI: electron impact GC: gas chromatography HCl: hydrochloric acid HFBI: heptafluorobutyryl imidazole MS: mass spectrometry PAO: phenylarsine oxide PDT: 1,3-propanedithiol SPE: solid phase extraction SPME: solid phase microextraction Data sources: (1) Logan TP, Smith JR, Jakubowski EM, Nielson RE. Verification of lewisite exposure by the analysis of 2-chlorovinyl arsonous acid in urine. Toxicol Meth. 1999;9:275–284. (2) Wooten JV, Ashley DL, Calafat AM. Quantitation of 2-chlorovinylarsonous acid in human urine by automated solid-phase microextraction-gas chromatography-mass spectrometry. J Chromatogr B Analyt Technol Biomed Life Sci. 2002;772:147–153. (3) Fidder A, Noort D, Hulst AG, de Jong LP, Benschop HP. Biomonitoring of exposure to lewisite based on adducts to haemoglobin. Arch Toxicol. 2000;74:207–214.

730 Medical Diagnostics

urine samples. The blood assay for both bound and free one animal study), but also indicates a substantial CVAA will potentially provide a longer opportunity decrease (90%) in concentration levels observed over for retrospective confirmation of exposure (based on a 10-day period.

CYANIDE

Background 5’-triphosphate. Other proteins and enzymes are also affected by cyanide, such as superoxide dismutase and Cyanide is an important industrial chemical that has xanthine oxidase, and may also contribute to cyanide’s many uses. It is produced in large quantities across the toxic effects.131 world and people are exposed to cyanide in a variety Cyanide’s metabolism in the body is very rapid and of ways. Many plants, including cassava roots, lima can occur at 0.017 mg/min/kg.131 The most common beans, and bamboo shoots, contain small amounts form of metabolism is the conversion to thiocyanate of cyanogenic glycosides.131–133 Polymers that contain (SCN-), which is then excreted via the kidneys.143,144 The carbon and nitrogen release cyanide when burned. mitochondrial enzyme rhodanese (thiosulfate sulfur For example, large amounts of cyanide can be released transferase) is thought to be the main catalyst for the during house and forest fires and from tobacco smoke formation of SCN-, but β-mercaptopyruvate-cyanide (100–1600 parts per million [ppm]).131 Some drugs also transferase can transform cyanide to SCN- through a contain cyanide or have active ingredients that are different route.141,145 The conversion to SCN- is thought converted to cyanide in the body, including sodium to be limited by the amount of thiosulfate. Cysteine, nitroprusside, which is given for the critical care of cystine, GSH, and β-mercaptopyruvic acid can also be hypertension.131 Synonyms for hydrogen cyanide sulfur sources.131, 141,145 The reaction with cystine is also (HCN) include prussic acid, formic anamminide, and an important pathway leading to 15% of the total cya- fromonitrile. Deaths due to cyanide intoxication are nide dose excreted in the urine as 2-iminothiozoline-4- usually due to the inhalation of HCN or the ingestion carboxylic acid.144 Other elimination pathways are the of cyanide salts. The signs of cyanide poisoning include exhalation of HCN and oxidation to cyanate and reac- 146 headache, nausea, vomiting, tachypnoea, dyspnoea to tions with vitamin B12 to form cyanocobalamin. convulsion, unconsciousness, and death.131, 134–137 Analytical Methods Cyanide in the Human Body Several matrices have been used to assess cyanide Cyanide, the CN- ion, exists as HCN in the body at exposure. Whole blood is the most common, but mea- physiological pH. The mechanism of cyanide toxicity surements have been made in serum, plasma, saliva, is believed to be the inactivation of iron III (ferric) tissues, gastric aspirate, and urine. enzymes in the body. The inhibition of cytochrome oxidase, which disrupts mitochondrial oxidative Whole Blood phosphorylation, is thought to be the most important mechanism for cyanide toxicity.131, 138–141 Cyanide also Whole blood has been the matrix of choice, thus far, binds to the hemoglobin in erythrocytes.142 The bind- to determine cyanide exposure in humans. However, ing of cyanide to iron III enzymes and proteins is there are problems associated with cyanide and SCN- reversible.140,141 measurements in blood. Sample collection, storage, At low concentrations, 93% to 99% of total cyanide and preparation are very important. Different levels is bound to methemaglobin (metHb) in the erytho- of cyanide have been found in samples collected from cytes.131 The metHb is the deoxygenated iron III form different vessels (venous, arterial, and ventricular). of hemoglobin. Cyanide has a very high affinity to Different whole blood samples can contain different metHb and a low affinity to the oxygenated form of amounts of metHb, which has a high affinity to cya- hemoglobin. Usually less than 1% of the total hemo- nide. Also, free HCN may be more important because globin is in the metHb form, so at increased concen- it is the form that reacts with cytochrome oxidase and trations of cyanide in the blood, a larger amount is causes the most significant adverse health effects.131 found in the serum. In tissue, cyanide binds to the Storage of the whole blood is also critical but not well heme group in mitochondrial cytochrome oxidase understood. Some studies have shown an increase in and inhibits electron transport.140,141 All tissues are af- cyanide of up to 40% upon storage of whole blood for fected by this enzymatic inhibition, but especially those 1 week and a 14% increase after 1 day at 4°C, while that require high amounts of oxygen and adenosine others have shown a decrease of 20% to 30% within 1

731 Medical Aspects of Chemical Warfare

Table 22-13 Analytical Methods for Determining Cyanide in Biological Samples1

Percent Sample Matrix Preparation Method Analysis Method LOD Recovery

Blood Separation in MDC; derivatization Spectrophotometry 0.1 ppm NR2 Blood Separation in MDC; derivatization Spectrofluorometry (total CN) 0.025 ppm NR3 Plasma Deproteination with TCA; Spectrophotometry (SCN-CN ~ 0.07 ppm 96 (SCN)4 derivatization determination) Erythrocyte Sample purged; absorption of Spectrophotometry (SCN-CN nR 93–975 suspension hcn in NaOH; oxidation of SCN determination) Blood cells centrifugation to separate cells; hplc-fluorescence detection 0.002 ppm 836 extraction; derivatization Blood Acidification Headspace GC-NPD ~ 0.03 ppm NR7 Blood Acidification; derivatization headspace GC-ECD 0.1 ppm NR8 Blood Separation in MDC; color Spectrophotometry ~ 0.07 ppm NR9 development Blood Incubation of acidified sample GC-NPD 0.001 ppm NR10 Blood Separation in MDC; absorption in Spectrophotometric (free CN) 0.4 ppm ~ 8011 methemoglobin Blood Acidification GC-NPD 0.014 ppm 86–9912 Blood Microdiffusion, derivatization isotope dilution LC-MS 5 ppb N/A13 Blood and liver Sample digestion; treatment with Specific ion electrode (total CN) 0.005 ppm 100–10914 lead acetate; absorption with NaOH Blood and urine Separation in MDC; derivatization Spectrofluorometric 0.008 ppm 66–83 (blood) 76–82 (urine)15 Urine Dilution; derivatization Spectrophotometry (SCN-CN ~ 0.07 ppm 88 (SCN)16 determination) Saliva Derivatization HPLC-UV (SCN) 2 ng (on instrument) 95–9917 Serum, urine, saliva extraction of buffered sample flame AAS (SCN) 0.004 ppm 96–10218 with isoamyl acetate Serum Addition of acetonitrile; Spectrophotometry (SCN) 0.3 ppm 9419 centrifugation; separation Urine, saliva Basify; derivatization; extraction; GC-ECD (SCN) ~ 0.033 83–10620 back extraction Urine, saliva dilution; filtration Ion chromatography-UV (SCN) 0.02 95–10121 Urine Ion chromatography; acidification; Spectrophotometry (SCN) ~ 0.145 ppm nR22 derivatization (lowest reported) Urine Dilution; solid phase extraction Suppressed ion chromato- ~ 0.011 ppm NR23 graphy with conductivity detection Urine (2-aminothio- cation exchange; reduction; Suppressed ion chromato- ~ 0.03 ppm NR24 zoline-4-carboxylic derivatization graphy with fluorescence acid) detection Urine (2-aminothio- Solid phase extraction and isotope dilution GC-MS 25 ppb N/A25 zoline-4-carboxylic derivatization acid)

(Table 22-13 continues)

732 Medical Diagnostics

Table 22-13 continued

AAS: atomic absorption spectrometry ATC: 2-amino-thiazoline-4-carboxylic acid CN: cyanide ECD: electron capture detection GC: gas chromatography GC-MSD: gas chromatography-mass selective detection HCN: hydrogen cyanide HPLC: high-performance liquid chromatography LC: liquid chromatography LOD: limit of detection MDC: microdiffusion cell NaOH: sodium hydroxide NPD: nitrogen phosphorus detection NR: not reported ppb: parts per billion ppm: parts per million SCN: thiocyanate TCA: trichloroacetic acid UV: ultraviolet absorbance detection Data sources: (1) US Department of Health and Human Services, Public Health Services, Agency for Toxic Substances and Disease Registry. Toxicological profile for cyanide. Atlanta, Ga: USDHHS; 1997. (2) Morgan RL, Way JL. Fluorometric determination of cyanide in biological fluids with pyrudixal. J Anal Toxicol. 1980;4:78–81. (3) Ganjeloo A, Isom GE, Morgan RL, Way JL. Fluorometric determination of cyanide in biological fluids with p-benzoquinone. 1980;55:103–107. (4) Pettigrew AR, Fell GS. Simplified colorimetric determination of thiocyanate in biological fluid, and its application to investigation of the toxic amblypias. Clin Chem. 1972;18:996–1000. (5) McMillan DE, Svoboda AC 4th. The role of erythrocytes in cyanide detoxification. J Pharmacol Exp Ther. 1982;221:37–42. (6) Sano A, Takimoto N, Takitani S. High- performance liquid chromatographic determination of cyanide in human red blood cells by pre-column fluorescence derivitization. J Chro- matogr. 1992;582:131–135. (7) Levin BC, Rechani PR, Gruman JL, et al. Analysis of carboxyhemoglobin and cyanide in blood of victims of the DuPont Plaza Hotel fire in Puerto Rico. J Forensic Sci. 1990;35:151–168. (8) Odoul M, Fouillet B, Nouri B, Chambon R, Chambon P. Specific determination of cyanide in blood by headspace gas chromatography. J Anal Toxicol.1994;18:205–207. (9) Laforge M, Buneaux F, Houeto P, Bourgeios F, Bourdon R, Levillain P, et al. A rapid spectrophotometric blood cyanide determination applicable to emergency toxicology. J Anal Toxicol.1994;18:173–175. (10) Seto Y, Tsunoda N, Ohta H, et al.Determination of blood cyanide by headspace gas chromatography with nitrogen phosphorus detection and using a megabore capillary column. Analytica Chimica Acta.1993;276:247–259. (11) Tomoda A, Hashimoto K. The determination of cyanide in water and biological tissues by methemoglobin. J Hazardous Materials. 1991;28:241–249. (12) Calafat AM, Stanfill SB. Rapid quantitation of cyanide in whole blood by automated headspace gas chromatography. J Chromatography B Analyt Technol Biomed Life Sci. 2002;772:131–137. (13) Tracqui A, Raul JS, Geraut A, Berthelon L, Ludes B. Determination of blood cyanide by HPLC-MS. J Anal Toxicol. 2002;26:144–148. (14) Egekeze JO, Oehme FW. Direct potentiometric method for the determination of cyanide in biological materials. J Anal Toxicol. 1979;3:119–124. (15) Sano A, Takezawa M, Takitani S. Spectrofluorometric determination of cyanide in blood and urine with naphthalene-2,3-dialdehyde and taurine. Anal Chim Acta. 1989;225:351–358. (16) Pettigrew AR, Fell GS. Simplified colorimetric determination of thiocyanate in biological fluid, and its application to investigation of the toxic amblyopias. Clin Chem. 1972;18:996–1000. (17) Liu X, Yun Z. High-performance liquid chromatographic determination of thiocyanate anions by derivatization with pentafluorobenzyl bromide. J Chromatogr A. 1993;653:348–353. (18) Chattaraj S, Das AK. Indirect determination of thiocyanate in biological fluids using atomic absorption spectrometry. Spectrochica. 1992;47:675–680. (19) Li HZ, Bai G, Sun RM, Du LK. Determination of thiocyanate metabolite of sodium nitroprusside in serum by spectrophotometry. Yao Xue Xue Bao. 1993:28:854–858. (20) Chen SH, Wu SM, Kou HS, Wu HL. Electron-capture gas chromatographic determination of cyanide, iodide, nitrite, sulfide and thiocyanate anions by phase-transfer-catalyzed derivatization with pentafluorobenzyl bromide. J Anal Toxicol. 1994;18:81–85. (21) Michigami Y, Fujii K, Ueda K, Yamamoto Y. Determination of thiocyanate in human saliva and urine by ion chromatography. Analyst. 1992;117:1855–1858. (22) Tominaga MY, Midio AF. Modified method for the determination of thiocyanate in urine by ion-exchange chromatography and spectrophotometry. Rev Farm Bioquim Univ Sao Paulo. 1991;27:100–105. (23) Miura Y, Koh T. Determination of thiocyanate in human urine samples by suppressed ion chromatography. Anal Sci 7. 1991;(Suppl Proc Int Congr Anal Sci, 1991, Pt 1):167–170. (24) Lundquist P, Kagedal B, Nilsson L, Rosling H. Analysis of the cyanide metabolite 2-aminothiazoline-4-carboxylic acid in urine by high-performance liquid chromatography. Anal Biochem. 1995;228:27–34. (25) Logue BA, Kirschten NP, Petrovics I, Moser MA, Rockwood GA, Baskin SI. Determination of the cyanide metabolite 2-aminothiazole-4-carboxylic acid in urine and plasma by gas chromatography-mass spectrometry. J Chromatogr B Analyt Technol Biomed Life Sci. 2005;819:237–244.

day at 4°C and continued decreases over 2 weeks. A Conway cell, although there have been some problems decrease in cyanide levels was seen when whole blood associated with the Conway cell. Other methods use was stored at 2°C. Results from storage at – 20°C has acid to release the cyanide followed by a variety of been disputed, with some studies showing up to 3.5 techniques. Most of these methods are colorimetric times the original cyanide levels and others showing no and rely on the König reaction to produce a dye that change.147 These differences appear to be independent is quantified using spectrophotometry. Cyanide and of the original cyanide concentrations.148 SCN- both react and must be separated using micro- Prior to detection, cyanide must first be separated diffusion or distillation. Additionally, most of these from hemoglobin. This separation is most often done methods are time consuming and lack specificity or with sulfuric acid and followed by microdiffusion in a sensitivity. Detection limits using this approach are

733 Medical Aspects of Chemical Warfare generally in the high parts-per-billion range. There with electron capture detection. Some urine methods have also been assays developed that acidify the have measured the urinary metabolite 2-aminothiozo- solution and then sample the head space for HCN line-4-carboxylic acid. followed by GC with a nitrogen-phosphorus detector or derivatization followed by GC with an electron Reference Range Values capture detector. Detection limits by nitrogen-phosphorus detection and electron capture detector have Reference range values for cyanide in various been in the high parts-per-billion range. Cyanide has matrices tend to vary greatly depending on the been measured in RBCs by high-performance LC after study and on the method of analysis. Thus, a refer- derivatization with fluorescence detection. ence range needs to be established for any method. In some of the studies, the range of blood cyanide Urine, Saliva, and Serum or Plasma levels in normal populations is less than 150 parts per billion and urinary SCN- less than 1.0 mg/mL.131 Methods to detect cyanide exposure in human urine, Smokers have much higher cyanide levels than saliva, and either serum or plasma have concentrated nonsmokers; in some smokers blood cyanide levels on SCN-. These methods include derivatization with as high as 500 ng/mL have been reported, which high-performance LC with ultraviolet detection, de- is 50 times higher than that typically reported for rivatization with spectrophotometric detection, or GC nonsmokers (Table 22-13).131

PHOSGENE

Background gene gas exposures can lead to pulmonary edema and death. Exposure to liquid phosgene by direct skin or Phosgene, also known as carbonyl chloride, was eye contact is rare but is thought to produce localized used extensively in World War I and caused more severe burns.156 Inhalation is the most toxic exposure deaths than any other agent.149 It is now a widely used route for phosgene and the route thought to have industrial chemical. In 2002 the global production of caused most of the injuries and deaths during World phosgene was estimated to be over 5 million metric War I. A patient who has inhaled phosgene requires tons,150 most of which is consumed at the production advanced treatment techniques. site.151 The first synthesis of phosgene was performed Phosgene is also a powerful acylating agent that in 1812 by exposing a mixture of chlorine and carbon reacts with nucleophiles such as amines, sulfides, or monoxide (CO) to sunlight.152 During World War I hydroxyls. Phosgene’s toxicity was originally thought phosgene was produced in bulk by the reaction of to be due to HCl generation during the reaction of chlorine and CO in the presence of activated carbon phosgene with moisture in the body. Later, acylation catalyst.152 Phosgene is generally produced in the reactions were found to be responsible for a majority same way today but with higher efficiency because of of phosgene’s toxic effects. Phosgene is greater than newer high-surface–area catalysts.153 Phosgene is an 800 times more toxic than HCl. Free amines protect important intermediate in many industrial products, against phosgene poisoning but not against the toxic including insecticides, isocyanates, plastics, dyes, and effects of HCl, phosgene inhibits coenzyme I and HCl resins.150 Additionally, phosgene is formed during does not, and chemically similar compounds (such the combustion of chlorinated hydrocarbons during as ketene) that do not have chlorine to generate HCl fires,154 and by the photooxidation of chlorinated sol- have similar toxicity to phosgene.157 Furthermore, vents in the atmosphere.155 phosgene’s rate of reaction with free amines has been At room temperature (20°C) phosgene is a fuming found to be much higher than its rate of reaction with liquid with a vapor pressure of 1180 mm Hg and boil- water. In a solution of aniline in water, phosgene reacts ing point of 7.6°C. The gas is heavier than air, with almost exclusively with the aniline.152 a relative density ratio of 4.39 at 20°C. This density Most of the data on health effects from phosgene are allows phosgene gas to collect in low-lying areas. from inhalation exposures. The regulatory threshold Phosgene’s odor is somewhat sweet and resembles that limit value (8-hour, time-weighted average) for phos- of fresh cut grass or hay. At higher concentrations the gene has been established by the National Institute for odor becomes pungent or burning and causes rapid Occupational Safety and Health, an institute within olfactory fatigue.156 the Centers for Disease Control and Prevention, as 0.1 Exposure to phosgene gas causes irritation to the ppm158; the 60-minute and 24-hour emergency expo- eyes, nose, throat, and respiratory tract. Higher phos- sure limits are 0.2 ppm and 0.02 ppm, respectively.155

734 Medical Diagnostics

The estimated LCt50 of phosgene concentration in air serves as both a scavenger of reactive compounds to related to exposure time through inhalation is 500 protect cells and as a store for cysteine moieties.162 GSH ppm/min.156 The aerosol exposure that is lethal to is found in general concentrations in healthy adults 162,163 100% of the exposed population (LCt100) is estimated in the millimolar range. GSH can be oxidized to to range from 1,300 ppm/min to 1,600 ppm/min.159 glutathione disulfide (GSSG), a simple dimer of GSH The nonlethal levels of phosgene are estimated to be joined by a disulfide linkage. The ratio of the GSH re- less than 300 ppm/min, and 25 ppm/min is regarded dox couple in vitro has been determined to be between as the threshold for lung damage.156 100:1 and 10:1 GSH to GSSG.162,163 When depleted, GSH Phosgene odor can be recognized at levels greater fails to protect cellular oxidation and leads to irrevers- than 1.5 ppm/min, with irritation in the mucus mem- ible oxidative damage.164 branes occurring at 3 ppm/min or higher.157 Exposure Phosgene was found to react with GSH to form limits that cause adverse health effects can be reached an acylated dimer, diglutathionyl dithiocarbonate either by longer exposure to lower concentration or (GSCOSG; Figure 22-13).165 This marker for phosgene shorter exposure to higher concentration. In one study, metabolism was first discovered by Pohl et al 165 in the however, workers exposed to daily phosgene concen- metabolism of chloroform in the liver, where phosgene trations above 1 ppm but less than 50 ppm showed was believed to be generated by enzymatic action from no difference in mortality or morbidity compared to chloroform. The bisglutathione adduct GSCOSG was workers in the same plant who were unexposed.160 detected in vivo in the bile of rats exposed to chloro- The concentrations of phosgene in air that cause form and in vitro in rat liver microsomes. Pohl et al also acute effects have been studied in many animal mod- found a decrease in GSH levels in the rats.165 GSCOSG els. A review of previous animal studies in the litera- was found directly by mixing phosgene with GSH in ture was performed by Diller and Zante to estimate the buffer solution.166 approximate inhalation-dose–toxicity relationship for The in-vivo generation of GSSG and CO has also 161 many species of animals. In this report, animal LCt50 been reported in the blood of mice exposed to halo- ranged from approximately 200 ppm/min for cats to forms.167 The observation of GSSG and CO may be 2000 ppm/min for goats. Guinea pigs and mice had linked to the further metabolism of GSCOSG after the same approximate value as humans, 500 ppm/min. generation, but there are no studies that have identi-

Dogs and rabbits had higher LCt50 values than humans fied this potential relationship. (1000 ppm/min and 1500 ppm/min, respectively), GSH is also found in tissue, so similar reactions with while nonhuman primates and rats had estimated GSH and phosgene are expected in the bronchoalveo- lower LCt50 values than humans (300 ppm/min and lar region. In the excised lung tissue from rabbits and 400 ppm/min, respectively).161 mice inhalationally exposed to phosgene, there was

Phosgene Metabolism and Markers for Phosgene Exposure HO

Phosgene is very reactive and is believed to be O

quickly transformed in vivo. Phosgene reacts with H2N amino, hydroxyl, and thiol groups. In the blood, it can react with a variety of proteins, including albumin HO NH2 and hemoglobin. It also reacts with GSH and cysteine. Because chloroform is metabolized to phosgene in O O the body, it is expected that low levels of phosgene’s O HN protein adducts and metabolites can be found in the NH general background population because of low-level S O NH incidental exposure to chloroform. Studies are needed O S to accurately determine this reference range. O NH O O Glutathione and Glutathione Adducts HO HO

Because phosgene is a highly reactive acylating Chemical Formula: C21H32N6O13S2 agent, it is expected to interact with the body’s anti- Molecular Weight: 640.64 oxidant defense system. GSH is a tripeptide thiol con- sisting of cysteine, glutamic acid, and glycine, which Fig. 22-13. Structure of diglutathionyl dithiocarbonate.

735 Medical Aspects of Chemical Warfare little change in tissue total GSH concentration, but have been approximated at 250 µm in plasma and 400 the reduced GSH levels fell significantly and GSSG µm in urine.174 levels increased significantly.168,169 In these studies, the The reaction of phosgene and cysteine in vitro was reduced GSH as percent of total was 41% less than the first reported by Kubic and Anders, and the product control subject. The collection of lung tissue samples was identified as 2-oxothiazolidine-4-carboxylic acid from humans that are thought to have been exposed (known as OTZ, OTC, or procysteine).172 OTZ was to phosgene is impractical, but the collection of bron- formed by incubating hepatic microsomal fractions choalveolar lavage fluid is possible. Sciuto has reported with chloroform, nicotinamide adenosine dinucleotide on the concentrations of GSSG, GSH, and total GSH in phosphate, and cysteine (Figure 22-14).172 Isotopic the bronchoalveolar lavage fluid of rodents with severe studies showed that chloroform was first metabolized changes in the GSH redox state following phosgene to phosgene by cytochrome P-450. The phosgene then inhalation.170,171 Increases in total GSH levels were ob- reacted with cysteine to form OTZ. Synthetic routes to served in both studies, but the redox state of the GSH OTZ also have indicated the generation from cysteine was not reported. It can be anticipated that increased and phosgene.175 levels of the oxidized form of GSH would be a better OTZ is a prodrug of cysteine and is rapidly con- indicator of phosgene exposure, but at this time only verted by 5-oxoprolinase to cysteine in vivo.173 Studies total GSH has been correlated with phosgene inhalation have indicated that the lifetime of OTZ in the human in lavage fluid. It should also be noted that in Sciuto’s body varies from 3 to 8 hours.173,176 OTZ’s rapid elimi- reported studies, there was no discrimination between nation limits its usefulness as a biomarker for phosgene GSSG and GSCOSG. Because these two species are simi- unless samples are obtained quickly after a suspected lar in structure, it is unclear whether the reported detec- exposure. tion of GSSG includes some component from GSCOSG. Other possible markers for phosgene exposure may The detection of GSCOSG, increased GSSG levels, be the reduction of cysteine to its disulfide, cystine, and decreased GSH levels following phosgene expo- upon exposure to phosgene. However, this reaction sure may hold promise for detecting phosgene expo- is not specific to phosgene exposure. The formation sure, but there are considerable obstacles to some of of the acylated dimer cysteine-CO-cysteine has been these approaches. The levels of total GSH and changes shown in vitro when cysteine is treated with phosgene in GSH redox state in animal models were determined in solution, and this reaction occurs even in the pres- shortly after exposure to phosgene; however, the life- ence of GSH.166 time of GSCOSG is not known, and even though the levels of total GSH in mice exposed to phosgene were Small Molecule Adducts significantly higher than in controls up to 24 hours postexposure, the GSH levels were only marginally Inhaled phosgene reportedly reacts with other com- above controls at times up to 7 days postexposure.171 ponents in the lungs. In 1933 treatment of lung pulp Furthermore, there are large variations in GSH and with phosgene was found to form the chlorocarbonic GSSH levels in human subjects, so predetermined base- ester of cholesterol.177 Kling 177 also indicated that the line levels should be required for each patient.162,163 The hydrophilic character of the fats was destroyed in the formation of GSCOGS may be specific for phosgene (or cells because of the loss of free sterols and suggested chlorinated compounds that metabolize to phosgene). that caused acute pulmonary edema. A reference range study of GSCOSG levels in people Phosgene forms adducts with phospholipids. In a with no known exposure to phosgene is required to use this biomarker for phosgene exposure.

O Cysteine Adducts

S Phosgene reportedly forms an adduct with the HN thiol amino acid cysteine.172 Cysteine is a building block of GSH and is the rate-limiting step in generat- O ing GSH for antioxidant tissue protection.173 As in the case of GSH, cysteine can be reduced to the disulfide OH cysteine dimer cystine. Cysteine is generated in the Chemical Formula: C4H5NO3S body by the conversion of methionine through the Molecular Weight: 147.15 cystathionine pathway. Average concentrations of cysteine in humans are typically lower than GSH and Fig. 22-14. Structure of 2-oxothiazolidine-4-carboxylic acid.

736 Medical Diagnostics

study of chloroform metabolism, a single phospholipid gene reacted with both albumin and hemoglobin and adduct was formed and it was thought to be related that there was a higher level of adduct formation to phosphatidylethanolamine (PE) because PE was the with the albumin than with the hemoglobin. The only phospholipid found to be depleted.178 In other analysis of the albumin-14C–labeled phosgene ad- studies on the metabolism of chloroform, a single duct indicated that the major site of adduct forma- phosgene adduct was found and identified as an ad- tion was an intramolecular lysine-lysine adduct with duct of PE with the phosgene carbonyl group bound to a CO bridge.186 Analysis of the hemoglobin-phosgene the amine of the head group of PE.179,180 Further work adduct showed the presence of a hydantoin func- on this PE adduct characterized it as a dimer of PE, the tion between the N-terminal valine and the amino two subsistents linked at the amine of the head group portion of leucine in a fragment containing amino by the carbonyl from phosgene.181 Furthermore, this acids 1 to 5.186 adduct has been implicated as the critical alteration Noort et al186 were able to detect the phosgene- leading to cell death by chloroform metabolism.182 albumin adduct in whole blood that was treated with Nonspecific increases in the phospholipid content of 1 μM phosgene and the phosgene-hemoglobin adduct lavage fluid have been observed in rats 6 hours after in whole blood treated with 1 mM phosgene. The exposure to inhalational phosgene, but specific adducts phosgene-albumin adduct was detectable at an order were not monitored.183 of phosgene concentration magnitude 3 times lower than the phosgene-hemoglobin adduct. The higher Macromolecule Adducts detection limits of the globin adduct were reportedly in part due to an interference of natural hydantoin Lung tissue damage after phosgene exposure, which function in the same position as the phosgene ad- leads to permeability of the blood-air barrier and pul- duct.186 Sample preparation for the more sensitive monary edema, is likely due, in part, to the reactions albumin-based method included isolating albumin of phosgene with various macromolecules, including by affinity chromatography, carboxymethylation, proteins. The permeability of the blood-air barrier also dialysis, and tryptic digestion. LC-MS-MS analysis raises the possibility of phosgene entering the blood was done using either a high-resolution MS instru- stream and forming adducts with blood proteins. Some ment or by multiple reaction monitoring on a triple work has been done developing methods for identify- quadrupole mass spectrometer. The multiple reaction ing and quantifying phosgene adducts with abundant monitoring experiment observed the fragmentation blood proteins. of the doubly charged ion at 861 daltons (da) to both The initial evidence that phosgene reacts and forms the 747.5 da and 773.6 da transitions. Noort et al an adduct with hemoglobin was from a study of the estimate that the method in vivo should be able to metabolism of chloroform. In this study, Pereira et detect a 320 mg/min/m3 exposure, assuming 10% al allowed chloroform to be metabolized with liver absorption of the dose by the blood, resulting in an microsomes.184 One of chloroform’s initial metabolites adduct concentration of 1.3 µm, just above the detec- is believed to be phosgene. The study authors identi- tion limit for the albumin adduct.186 fied N-hydroxymethyl cysteine by GC-MS, resulting from the reaction of phosgene of cysteine moiety in Other Indicators of Phosgene Exposure hemoglobin.184 Additionally, Fabrizi et al identified phosgene adducts with petapeptides of lysozyme and Inhalational exposure to phosgene causes severe the N-terminal peptide from human histone H2B, even inflammation and tissue damage. There are many when these peptides were treated with phosgene in indicators for the biochemical changes caused by solution in the presence of GSH, a known phosgene phosgene exposure that measure the biochemical scavenger.168 There is evidence that phosgene reacts changes in the tissues that are damaged by phosgene. and forms blood protein adducts in vivo. Sciuto et al re- These types of markers are not as specific for phos- ported spectrophotometric differences in plasma from gene exposure as small molecule or protein adduct mice, rats, and guinea pigs exposed to high doses of markers, but they can suggest that an individual has phosgene.185 This study also suggested that phosgene been exposed to phosgene and indicate the severity could directly attack RBCs after observing the shifted of the exposure. fragility curve for erythrocytes.185 Some of the most important nonspecific indicators Noort et al have developed methods to detect and for phosgene-mediated tissue damage are associ- quantify phosgene adducts with both hemoglobin ated with inflammation. Sciuto et al reported that and albumin.186 Treatment of whole blood with ra- interleukin (IL)-6 levels were highly elevated in the dioisotopically labeled phosgene showed that phos- bronchoalveolar lavage fluid from rodents exposed

737 Medical Aspects of Chemical Warfare to phosgene compared to levels in controls.185 Levels expected that enzymes responsible for maintaining of IL-6 were 16-fold higher 4 hours after the phosgene GSH levels may also be affected, and their levels can exposure, peaked at 12 hours, and were still over 100- be monitored to suggest phosgene exposure. Sciuto fold higher than in controls 72 hours after phosgene et al found levels of GSH peroxidase and GSH re- exposure. Macrophage inflammatory protein was also ductase significantly increased in the bronchoalveolar affected in the rodents exposed to phosgene, showing a lavage fluid of mice exposed to 160 to 220 ppm/min 10-fold increase compared to the control rodents 8 and phosgene.171 The levels of GSH peroxidase and GSH 10 hours after the exposure, but macrophage inflam- reductase were measured by ELISA and were elevated matory protein levels returned to near-normal levels up to 72 hours after exposure.171 24 hours after the phosgene exposure.185 Sciuto et al The amount of total protein in bronchoalveolar reported that changes in IL-4, IL-10, tumor necrosis lavage fluid may also be a marker for the extent factor, and IL-1β levels were minimal after phosgene of lung damage after a phosgene exposure. Sciuto exposure as compared to changes in IL-6 levels.185 found that levels of total protein in the lavage fluid Sciuto et al also studied changes in fluid elec- of mice was dramatically increased following phos- trolyte levels in blood and bronchoalveolar lavage gene exposure and remained significantly elevated fluid from mice following exposure to phosgene.187 up to 7 days postexposure. Other reports by Sciuto Blood levels of sodium, chloride, and calcium were have indicated similar increases in protein levels in unchanged after exposure to phosgene compared to bronchoalveolar lavage fluid and have suggested controls, but the levels of potassium increased dras- they are due to increased permeability of the blood- tically and approached physiologically dangerous air barrier.170 In an earlier study, Currie et al found levels. In the bronchoalveolar lavage fluid, chloride significant increases in the concentration of bron- levels were unchanged, sodium dropped between 4 choalveolar lavage fluid protein immediately after and 12 hours, and both potassium and calcium levels a relatively low exposure (60 ppm/min).188 Total increased significantly after 4 hours and by a factor protein levels in bronchoalveolar lavage fluid are not of 3 after 8 hours, compared to nonexposed controls. specific for phosgene exposure, but may be helpful The simplicity of determining electrolyte levels offers to determine the extent of lung damage caused by promise for suggesting exposure to phosgene, but is it. A simple spectrophotometric assay is suitable not specific for phosgene exposure and requires sup- to determine protein levels because the amount of porting analyses. protein in lavage fluid is nonspecific and measures Because phosgene reacts rapidly with GSH, it is only total protein levels.

3-QUINUCLIDINYL BENZILATE

Background by an oral route is about 80% that of intravenous or intramuscular routes. Under optimum conditions, Commonly termed “QNB,” 3-quinuclidinyl ben- BZ is also about 40% to 50% as effective by inhalation zilate (BZ) is an anticholinergic glycolate that has been as by injection. Initial symptoms typically occur 30 designated “agent BZ” by the military. BZ is the only minutes to 4 hours after exposure. Full recovery may known incapacitating agent that has ever been weap- require 3 to 4 days. onized for use by the US military. That occurred in the BZ has a molecular weight of 337 g/mol and a early 1960s when BZ was produced at the Pine Bluff melting point of 168°C. It is solid at room temperature Arsenal between 1962 and 1965. BZ was subsequently and has a negligible vapor pressure. BZ is relatively dropped from the chemical arsenal for several reasons, stable and moderately resistant to air oxidation and including concerns about variable and unpredictable moderate temperatures, and it undergoes hydrolysis effects. in aqueous solution to produce benzylic acid and 3- BZ’s action is very similar to other anticholinergics quinuclidinol. such as atropine and scopolamine, differing mainly in potency and duration of effect. The lethal dose of 3-Quinuclidinyl Benzilate in the Human Body an infectious organism required to produce infection in 50% of the population (ID50) for BZ is about 6.2 Information on BZ in the body is limited. It can take µg/kg (about 500 ng/person). BZ is about 25 times 3 to 4 days before symptoms of BZ intoxication resolve. as potent as atropine, but only about 3-fold more Apparently, most BZ is excreted by the kidneys and active than scopolamine. However, BZ’s duration urine is the preferred analysis matrix. Benzylic acid of action is typically much longer, and the uptake and 3-quinuclidinol are the probable main metabolites

738 Medical Diagnostics

N O (Figure 22-15). It has been estimated that 3% of BZ in OH the body is excreted as the parent compound. C O C Analytical Methods

There are few references to the analysis of BZ in humans. BZ is frequently used by neuropharma- O 3-Quinuclidinyl Benzilate cologists as a marker, but these methods are often not HO C quantitative. In 1988 the United States began work on C OH demilitarization of BZ stockpiles. As part of that work, N the National Institute of Standards and Technology and the US Army Research and Development labora- OH tory at Fort Detrick, Maryland, developed a method for monitoring workers. The method was based on 3-Quinuclidinol Benzylic Acid GC-MS of the trimethylsilyl (TMS) derivatives and monitored all three analytes (BZ, benzylic acid, and Fig. 22-15. Structure of 3-quinuclidinyl benzilate and the 3-quinuclidinol). Detection limits were 0.5 ng/mL for main metabolites 3-quinuclidinol and benzylic acid. BZ and 5 ng/mL for the hydrolysis products.189

SAMPLE CONSIDERATIONS

General withstand freezing temperatures without splitting. A minimum of 25 to 30 mL should be collected. Urine The following section describes the field collec- samples should be frozen immediately (– 70°C or dry tion, shipment, and storage of biomedical samples ice is preferred). according to the guidance from the Centers for Disease Control and the United States Army Medi- Whole Blood cal Research Institute of Chemical Defense. In most instances, blood (serum, plasma) and urine are the It is recommended that samples of whole blood most commonly collected samples. Samples to be be handled cautiously from the start of the collection collected depend upon the specific method required to maintain integrity and preclude the possibility of by the assay. The information provided in this section contamination, tampering, or mislabeling. All samples is intended to provide general guidelines for sample should be collected under the close supervision of a collection, shipment, and storage and is not intended healthcare provider or physician and witnessed by to be comprehensive. In all cases, questions regarding an unbiased observer if possible. Samples should be specific procedures to obtain and ship samples should obtained as soon as possible following the suspected be directed to the receiving laboratory by calling the exposure. Additional samples may be obtained for fol- laboratory or consulting its Web site before collecting low up. Blood should be collected in 5- or 7-mL blood the sample. tubes. Specific methods may require specific types of tubes, such as purple-top (ethylenediamine tetraace- Urine tic acid) tubes for plasma or the red-top, unopened, vacuum-fill tubes for serum. The collection of urine samples should be done For methods that analyze serum or plasma, it under the close supervision of a healthcare provider is useful to process whole blood samples by cen- or an unbiased observer to preclude the possibility trifugation followed by separation of the plasma of sample tampering. Care should be taken to ensure or serum from the RBC pellet. The plasma or se- appropriate handling so as to minimize the chances rum components can then be frozen (– 70°C or dry for contamination from the environment or handling ice is preferred) and stored or shipped on dry ice. personnel. The urine should be collected immediately When on-site blood processing is not convenient, following the suspected exposure or at the earliest pos- unprocessed whole blood samples should be stored sible time. A midstream urine collection is desirable. If immediately at 4°C. Packed RBCs may be stored at follow up is anticipated, additional samples should be 4°C or frozen, depending on the properties of the obtained at 24 hours. Urine should be collected in clean analyte and the needs of the analytical method to urine cups or screw-capped plastic containers that can be performed.

739 Medical Aspects of Chemical Warfare

Preparing, Shipping, and Storing Specimens as whole blood, add a layer of frozen cold packs and place the secondary containers on top of the cold packs. Care must be taken not only when gathering sam- Place additional cold packs or absorbent material be- ples for testing but when preparing them for transport. tween the secondary containers to reduce movement What follows are the basic guidelines for readying, within the container. Lastly, place a layer of frozen transporting, and storing samples. cold packs on top of the secondary containers. When shipping frozen samples (plasma, serum, urine), add Labeling a layer of dry ice on top of the cushioning material in the bottom of the shipping container. Do not use large Specimens should be labeled in accordance with the chunks of dry ice for shipment because they could shat- standing operating procedures of the receiving labo- ter items during transport. Place additional absorbent ratory to ensure forensic integrity. The labels should material between wrapped urine cups to reduce their be as comprehensive as possible and double-checked movement within the outer container. Finally, add an for accuracy. Label samples with the facility of origin additional layer of dry ice. and clear markings resistant to water and refrigeration or freezing conditions. Labels should also include Documentation (Shipping Manifest, Incident Re- patient identification information (name or other spe- port, Chain-of-Custody) cific identifiers), date and time of collection, specimen identity, and some identification of the collector. A Prepare separate documentation for each container list of samples with the corresponding names of indi- shipped. Prepare and place a shipping manifest des- viduals should be maintained at the facility of origin ignating sample identification numbers, quantity, and and should also be included with the samples if they type in a zippered plastic bag on top of the specimens are shipped. Wrap each sample top with waterproof, before closing and sealing the container. Maintain a tamper-evident, forensic evidence tape, being careful copy of the manifest at the point of origin. Enclose not to cover the sample identification labels. an incident report form with as much information as possible describing the incident, providing the date Packaging and time of suspected exposure, detailing the onset and description of symptoms, sample collection For blood, separate each tube from the others or time, and suspected agent involved. Include patient wrap individually to prevent direct contact. Tubes information, such as name, social security number, should be placed in secondary packages, such as a age, and gender, as well as a point of contact. The divided box wrapped with absorbent material and incident report form should be stored in a zippered sealed inside a plastic bag, other sealable containers, plastic bag on top of the specimens before closing and or individually wrapped tubes sealed inside a plastic sealing the container. A chain-of-custody form must bag. Place absorbent material between the primary also be included. Prepare a separate chain-of-custody receptacle and the secondary packaging. Use enough form for samples in each container. Indicate sample absorbent material to absorb the entire contents of identity, any pertinent descriptors, and the number primary receptacles. To facilitate processing and iden- of samples. Place the completed chain-of-custody tification, package blood tubes so that similar tubes are forms in a plastic zippered bag on the outside of the packaged together (eg, all purple tops together). For shipping container. urine, wrap frozen cups with absorbent material and place them into sealable secondary packaging, such Shipping as those described for blood. Do not ship frozen urine and blood in the same package. Close and secure the outer container with filamen- tous shipping or strapping tape. Affix labels and mark- Shipping Container ings. Place a label on the outer container that indicates the proper name, “Diagnostic Specimens.” For those The shipping container should be a sealable, poly- containers with dry ice, place a Class 9 (dry ice) label styrene foam or other insulated container capable of on the outer container. This label must indicate the maintaining the contents at the preferred temperature amount of dry ice in the container, the address of the for the specimens. For cushioning, place additional ab- shipper, and the address of the recipient. This label sorbent material in the bottom of the outer containers. must be placed on the same side of the container as For samples that require refrigeration conditions, such the “Diagnostic Specimens” label.

740 Medical Diagnostics

Storage should be in accordance with conditions dictated by the sample type. Blood should be stored refrigerated Upon arrival, the receiving laboratory should main- at 4°C. Plasma or serum should be stored frozen at tain a proper chain of custody. If the samples are not –70°C. RBCs can be stored refrigerated at 4°C or frozen processed immediately, they should be stored as soon at –70°C; freezing is preferred for long-term storage. as possible after arriving at the receiving laboratory. Avoid repeated cycles that move samples from frozen Storing samples either before or after they are shipped to thawed or refrigerated to room temperature.

summary

The general class of agents involved in severe intoxi- ase analysis, instrumentation typically involves MS cation (ie, OP nerve agents, vesicants, etc) can often be systems with either GC or LC techniques to separate recognized by symptom presentation. However, test- the analyte from other matrix components. Although ing is necessary to identify the specific agent involved. the methods are desirable because they afford a high In cases where poisoning is suspected at low levels and level of confidence for identifying the analyte, they symptoms do not clearly indicate intoxication with a are time and labor intensive. Consequently the turn- specific chemical warfare agent, testing can provide around time for analyses is greater than that expected additional information to help consider or rule out an for standard clinical tests. exposure. In general, confirmatory analyses should Chemical warfare agents have been used against not be initiated in the absence of other information both military and civilian populations. In many of that suggests a potential exposure has taken place; these cases, healthcare providers have learned that other evidence, such as patient signs or symptoms, it is critical to rapidly identify exposed personnel environmental monitoring and testing, and threat in- to facilitate appropriate medical treatment and sup- telligence information should also be considered. This port. Incidents involving large numbers of personnel information should be used to guide decisions about have shown that is also important to determine those what agent or class of agents should be the focus of not exposed to avoid unnecessary psychological testing and it should ultimately be used in conjunc- stress and overburdening the medical system. In tion with test results to determine whether or not an addition to medical issues, the political and legal exposure has occurred. ramifications of chemical agent use by rogue na- Analytical methods for verifying chemical agent ex- tions or terrorist organizations can be devastating. posure do not employ instrumentation that is routinely Therefore, it is important that accurate and sensitive used for standard clinical testing, such as automated analytical techniques be employed and appropri- clinical analyzers. With the exception of cholinester- ately interpreted.

REFERENCES

1. matsuda Y, Nagao M, Takatori T, et al. Detection of the sarin hydrolysis product in formalin-fixed brain tissues of victims of the Tokyo subway terrorist attack. Toxicol Appl Pharmacol. 1998;150:310–320.

2. department of the US Army. Assay Techniques for Detection of Exposure to Sulfur Mustard, Cholinesterase Inhibitors, Sarin, Soman, GF, and Cyanide. Washington, DC: DA; 1996. Technical Bulletin Medical 296.

3. taylor P. Anticholinesterase agents. In: Brunton LL, Lazo JS, Parker KL, eds. Goodman and Gilman’s the Pharmacological Basis of Therapeutics. 11th ed. New York, NY: McGraw-Hill; 2006:201–216.

4. Sidell FR. Nerve agents. In: Sidell FR, Takafuji ET, Franz DR, eds. Medical Aspects of Chemical and Biological Warfare. In: Zajtchuk R, Bellamy RF, eds. The Textbooks of Military Medicine. Part I. Washington, DC: Borden Institute; 1997:129– 179.

5. harris R, Paxman J. A Higher Form of Killing: the Secret Story of Chemical and Biological Warfare. New York, NY: Hill and Wang; 1982: 53.

6. degenhardt CE, Pleijsier K, van der Schans MJ, et al. Improvements of the fluoride reactivation method for the verification of nerve agent exposure. J Anal Toxicol. 2004;28:364–371.

741 Medical Aspects of Chemical Warfare

7. noort D, Benschop HP, Black RM. Biomonitoring of exposure to chemical warfare agents: a review. Tox Appl Pharmacol. 2002;184:116–126.

8. Benschop HP, De Jong LP. Toxicokinetics of soman: species variation and stereospecificity in elimination pathways. Neurosci Biobehav Rev. 1991;15:73–77.

9. Benschop HP, de Jong LP. Toxicokinetics of nerve agents. In: Somani SM, Romano JA, eds. Chemical Warfare Agents: Toxicity at Low Levels. Boca Raton, Fla: CRC Press; 2000:25–81. Chapter 2.

10. Van der Schans MJ, Lander BJ, van der Wiel, H, Langenberg JP, Benschop HP. Toxicokinetics of the nerve agent (±) -VX in anesthetized and atropinized hairless guinea pigs and marmosets after intravenous and percutaneous administration. Tox Appl Pharmacol. 2003;191:48–62.

11. Bonierbale E, Debordes L, Coppet L. Application of capillary gas chromatography to the study of hydrolysis of the nerve agent VX in rat plasma. J Chromatogr B Biomed Sci Appl. 1997;688:255–264.

12. harris LW, Braswell LM, Fleisher JP, Cliff WJ. Metabolites of pinacolyl methylphosphonofluoridate (soman) after enzymatic hydrolysis in vitro. Biochem Pharmacol. 1964;13:1129–1136.

13. polak RL, Cohen EM. The influence of oximes on the distribution of 32P in the body of the rat after injection of 32P- sarin. Biochem Pharmacol. 1970;19:865–876.

14. Reynolds ML, Little PJ, Thomas BF, Bagley RB, Martin BR. Relationship between the biodisposition of [3H]soman and its pharmacological effects in mice. Toxicol Appl Pharmacol. 1985;80:409–420.

15. little PJ, Reynolds ML, Bowman ER, Martin BR. Tissue disposition of [3H]sarin and its metabolites in mice. Toxicol Appl Pharmacol. 1986;83:412–419.

16. Shih ML, Smith JR, McMonagle JD, Dolzine TW, Gresham VC. Detection of metabolites of toxic alkylmethylphosphonates in biological samples. Biol Mass Spectrom. 1991;20:717–723.

17. Shih ML, McMonagle JD, Dolzine TW, Gresham VC. Metabolite pharmacokinetics of soman, sarin, and GF in rats and biological monitoring of exposure to toxic organophosphorus agents. J Appl Toxicol. 1994:14:195–199.

18. tsuchihashi H, Katagi M, Nishikawa M, Tatsuno M. Identification of metabolites of nerve agent VX in serum collected from a victim. J Anal Toxicol. 1998;22:383–388.

19. fredriksson SA, Hammarström LG, Henriksson L, Lakso HA. Trace determination of alkyl methylphosphonic acids in environmental and biological samples using gas chromatography/negative-ion chemical ionization mass spectrometry and tandem mass spectrometry. J Mass Spectrom. 1995;30:1133–1143.

20. miki A, Katagi M, Tsuchihashi H, Yamashita M. Determination of alkylmethylphosphonic acids, the main metabolites of organophosphorus nerve agents, in biofluids by gas chromatography-mass spectrometry and liquid-liquid-solid- phase-transfer-catalyzed pentafluorobenzylation. J Anal Toxicol. 1999;23:86–93.

21. driskell WJ, Shih M, Needham LL, Barr DB. Quantitation of organophosphorus nerve agent metabolites in human urine using isotope dilution gas chromatography-tandem mass spectrometry. J Anal Toxicol. 2002;26:6–10.

22. Barr JR, Driskell WJ, Aston LS, Martinez RA. Quantitation of metabolites of the nerve agents sarin, soman, cyclosarin, VX, and Russian VX in human urine using isotope-dilution gas chromatography-tandem mass spectrometry. J Anal Toxicol. 2004;28:371–378.

23. minami M, Hui DM, Katsumata M, Inagaki H, Boulet CA. Method for the analysis of methylphosphonic acid metabolites of sarin and its ethanol-substituted analogue in urine as applied to the victims of the Tokyo sarin disaster. J Chromatogr B Biomed Sci Appl. 1997;695:237–244.

24. nakajima T, Sasaki K, Ozawa H, Sekjima Y, Morita H, Fukushima Y, Yanagisawa N. Urinary metabolites of sarin in a patient of the Matsumoto incident. Arch Toxicol. 1998;72:601–603.

742 Medical Diagnostics

25. noort D, Hulst AG, Platenburg DH, Polhuijs M, Benschop H. Quantitative analysis of O-isopropyl methylphosphonic acid in serum samples of Japanese citizens allegedly exposed to sarin: estimation of internal dosage. Arch Toxicol. 1998;72:671–675.

26. Katagi M, Nishikawa M, Tatsuno M, Tsuchihashi H. Determination of the main hydrolysis products of organophosphorus nerve agents, methylphosphonic acids, in human serum by indirect photometric detection ion chromatography. J Chromatogr B Biomed Sci Appl. 1997;698:81–88.

27. Suzuki T, Morita H, Ono K, Maekawa K, Nagai R, Yazaki Y. Sarin poisoning in Tokyo subway. Lancet. 1995;345:980.

28. polhuijs M, Langenberg JP, Noort D, Hulst AG, Benschop HP. Retrospective detection of exposure to organophosphates: analyses in blood of human beings and rhesus monkeys. In: Sohns T, Voicu VA, eds. NBC risks: Current Capabilities and Future Perspectives for Protection. Dordrecht, Holland, the Netherlands: Kluwer Academic Publishers; 1999:513–521.

29. polhuijs M, Langenberg JP, Benschop HP. New method for retrospective detection of exposure to organophosphorus anticholinesterases: application to alleged sarin victims of Japanese terrorists. Toxicol Appl Pharmacol. 1997;146:156– 161.

30. heilbronn E. The effect of sodium fluoride on sarin inhibited blood cholinesterase. Acta Chem Scand. 1964;18:2410.

31. heilbronn E. Action of fluoride on cholinesterase. II. In vitro reactivation of cholinesterases inhibited by organophosphorus compounds. Biochem Pharmacol. 1965;14:1363–1373.

32. albanus L, Heilbeonn E, Sundwall A. Antidote effect of sodium fluoride in organophosphorus anticholinesterase poisoning. Biochem Pharmacol. 1965;14:1375–1381.

33. Jakubowski EM, Heykamp LS, Durst HD, Thompson SA. Preliminary studies in the formation of ethyl methylphosphonofluoridate from rat and human serum exposed to VX and treated with fluoride ion. Anal Lett. 2001;34:727–737.

34. Jakubowski EM, McGuire JM, Evans RA, et al. Quantitation of fluoride ion released sarin in red blood cell samples by gas chromatography-chemical ionization mass spectrometry using isoptope dilution and large-volume injection. J Anal Toxicol. 2004;28:357–363.

35. adams TK, Capacio BR, Smith JR, Whalley CE, Korte WD. The application of the fluoride reactivation process to the detection of sarin and soman nerve agent exposures in biological samples. Drug Chem Toxicol. 2004;27:77–91.

36. Black RM, Harrison JM, Read RW. The interaction of sarin and soman with plasma proteins: the identification of a novel phosphonylation site. Arch Toxicol. 1999;73:123–126.

37. deJong LP, van Dijk C. Formation of soman (1,2,2-trimethylpropyl-methylphosphonofluoridate) via fluoride-induced reactivation of soman-inhibited aliesterase in rat plasma. Biochem Pharmacol. 1984;33:663–669.

38. clement JG. Importance of aliesterase as a detoxification mechanism for soman (pinacolyl methylphosphonofluoridate) in mice. Biochem Pharmacol. 1984;33:3807–3811.

39. clement JG, Benschop HP, DeJong LP, Wolthuis OL. Stereoisomers of soman (pinacolyl methylphosphonofluoridate): inhibition of serum carboxylic ester hydrolase and potentiation of their toxicity by CBDP (2-2-methylphenoxy)-4H- 1,3,2-benzodioxaphosphorin-2-oxide) in mice. Toxicol Appl Pharmacol. 1987;89:141–143.

40. maxwell DM, Brecht KM, O’Neill BL. The effect of carboxylesterase inhibition on interspecies differences in soman toxicity. Toxicol Lett. 1987;39:35–42.

41. maxwell DM, Brecht KM. Carboxylesterase: specificity and spontaneous reactivation of an endogenous scavenger for organophosphorus compounds. J Appl Toxicol. 2001;21(suppl 1):S103–S107.

42. nagao M, Takatori T, Matsuda Y, et al. Detection of sarin hydrolysis products from sarin-like organophosphorus agent-exposed human erythrocytes. J Chromatogr B Biomed Sci Appl. 1997;701:9–17.

743 Medical Aspects of Chemical Warfare

43. nagao M, Takatori T, Matsuda Y, Nakajima M, Iwase H, Iwadate K. Definitive evidence for the acute sarin poisoning diagnosis in the Tokyo subway. Toxicol Appl Pharmacol. 1997;144:198–203.

44. fidder A, Hulst AG, Noort D, et al. Retrospective detection of exposure to organophosphorus anti-cholinesterases: mass spectrometric analysis of phosphylated human butyrylcholinesterase. Chem Res Toxicol. 2002;15:582–590.

45. noort D, Fidder A, van der Schans MJ, Hulst AG. Verification of exposure to organophosphates: generic mass spectrometric method for detection of human butyrylcholinesterase adducts. Anal Chem. 2006;78:6640–6644.

46. Silver A. The Biology of Cholinesterases. Amsterdam, North Holland, the Netherlands: North Holland Publishing Com- pany; 1974:58–67.

47. mortensen SR, Brimijoin S, Hooper MJ, Padilla S. Comparison of the in vitro sensitivity of rat acetylcholinesterase to

chlorpyrifos-oxon: what do tissue IC50 values represent? Toxicol Appl Pharmacol. 1998;148:46–49.

48. Wilson BW, Sanborn JR, O’Malley MA, Henderson JD, Billitti JR. Monitoring the pesticide-exposed worker. Occup Med. 1997;12(2):347–363.

49. nigg HN, Knaak JB. Blood cholinesterases as human biomarkers of organophosphorus pesticide exposure. Rev Environ Contam Toxicol. 2000;163:29–111.

50. lotti M. Cholinesterase inhibition: complexities in interpretation. Clin Chem. 1995;41:1814–1818.

51. Sidell FR. Clinical considerations in nerve agent intoxication. In: Somani SM, ed. Chemical Warfare Agents. San Diego, Calif: Academic Press; 1992:163.

52. Wolthuis OL, Benschop HP, Berends F. Persistence of the anti-cholinesterase soman in rats; antagonism with a non- toxic simulator of this organophosphate. Eur J Pharmacol. 1981;69:379–383.

53. Worek F, Eyer P, Szinicz L. Inhibition, reactivation and aging kinetics of cyclohexylmethylphosphonofluoridate-inhibited human cholinesterases. Arch Toxicol. 1998;72:580–587.

54. lockridge O, Masson P. Pesticides and susceptible populations: people with butyrylcholinesterase genetic variants may be at risk. Neurotoxicology. 2000;21:113–126.

55. Wilson BW, Henderson JD. Blood esterase determinations as markers of exposure. Rev Environ Contam Toxicol. 1992;128:55–69.

56. St Omer VEV, Rottinghaus GE. Biochemical determination of cholinesterase activity in biological fluids and tissues. In: Ballantyne B, Marrs TC, eds. Clinical and Experimental Toxicology of Organophosphates and Carbamates. Oxford, Ox- fordshire, England: Butterworth & Heinemann; 1992: 15–27.

57. ellman GL, Courtney KD, Anders V Jr, Feather-stone RM. A new and rapid colorimetric determination of acetylcholinesterase activity. Biochem Pharmacol. 1961;7:88–95.

58. Worek F, Mast U, Kiderlen D, Diepold C, Eyer P. Improved determination of acetylcholinesterase activity in human whole blood. Clin Chem Acta. 1999;288:73–90.

59. eyer P, Worek F, Kiderlen D,et al. Molar absorption coefficients for the reduced Ellman reagent: reassessment. Anal Biochem. 2003;312:224–227.

60. doctor BP, Toker L, Roth E, Silman I. Microtiter assay for acetylcholinesterase. Anal Biochem. 1987;166:399–403.

61. Walter Reed Army Institute of Research. Whole Blood Assay. US patent 6,746,850. June 8, 2004.

62. Gordon RK, Haigh JR, Garcia GE, et al. Oral administration of pyridostigmine bromide and huperzine A protects human whole blood cholinesterases from ex vivo exposure to soman. Chemico Biol Interact. 2005;157–158:239–246.

744 Medical Diagnostics

63. eQM Research, Inc. Test-mate ChE Cholinesterase Test System. Web site. Available at http://www.eqmresearch.com/. Accessed January 19, 2007.

64. taylor PW, Lukey BJ, Clark CR, Lee RB, Roussel RR. Field verification of Test-mate ChE. Mil Med. 2003;168:314–319.

65. ellin RI, Burkhardt BH, Hart RD. A time-modified method for measuring red blood cell cholinesterase activity. Arch Environ Health. 1973;27:48–49.

66. michel HO. An electrometric method for the determination of red blood cell and plasma cholinesterase activity. J Lab Clinical Med. 1949;34:1564–1568.

67. uS Department of Defense Cholinesterase Reference Laboratory. Standard operating procedure for the seventeen minute red blood cell acetylcholinesterase procedure. US Army Center for Health Promotion and Preventative Medicine: Aberdeen Proving Ground, Md; 2002. SOP CRL 40-2.7.

68. uS Department of Defense Cholinesterase Monitoring Program, Cholinesterase Reference Laboratory. Web site. Avail- able at http://chppm-www.apgea.army.mil/dls/CHOLIN.HTM. Accessed January 19, 2007.

69. dunn MA, Hackley BE, Sidell, FR. Pretreatment for nerve agent exposure. In: Sidell FR, Takafuji ET, Franz DR, eds. Medical Aspects of Chemical and Biological Warfare. In: Zajtchuk R, Bellamy RF, eds. The Textbooks of Military Medicine. Part I. Washington, DC: Department of the Army, Office of The Surgeon General, Borden Institute; 1997: 181–196.

70. triggle DJ, Mitchell JM, Filler R. The pharmacology of physostigmine. CNS Drug Rev. 1998;4:87–136.

71. uS Department of Health and Human Services. US Food and Drug Administration Web site. FDA approves pyridostigmine bromide as pretreatment against nerve gas. FDA News. PO3-08, February 5, 2003. Available at http://www.fda. gov/bbs/topics/NEWS/2003/NEW00870.html. Accessed: January 19, 2007.

72. Grunwald J, Raveh L, Doctor BP, Ashani Y. Huperzine A as a pretreatment candidate drug against nerve agent toxicity. Life Sci. 1994;54:991–997.

73. Jiang H, Luo X, Bai D. Progress in clinical, pharmacological, chemical and structural biological studies of huperzine A: a drug of traditional medicine origin for the treatment of Alzheimer’s disease. Curr Med Chem. 2003;10:2231–2252.

74. lallement G, Baille V, Baubichon D, et al. Review of the value of huperzine as pretreatment of organophosphate poisoning. Neurotoxicology. 2002;23:1–5.

75. Wang R, Yan H, Tang XC. Progress in studies of huperzine A, a natural cholinesterase inhibitor from Chinese herbal medicine. Acta Pharmacol Sin. 2006;27:1–26.

76. Raveh L, Grunwald J, Marcus D, Papier Y, Cohen, E, Ashani Y. Human butyrylcholinesterase as a general prophylactic antidote for nerve agent toxicity. In vitro and in vivo quantitative characterization. Biochem Pharmacol. 1993;45:2465– 2474.

77. yanagisawa N, Morita H, Nakajima T. Sarin experiences in Japan: acute toxicity and long-term effects. J Neurol Sci. 2006;249:76–85.

78. papirmeister B, Feister AJ, Robinson SI, Ford RD. Medical Defense Against Mustard Gas: Toxic Mechanisms and Pharma- cological Implications, Boca Raton, Fla: CRC Press; 1991: 91–122.

79. Black RM, Noort D. Methods for the retrospective detection of exposure to toxic scheduled chemicals. Part A: analysis of free metabolites. In: Mesilaakso M, ed. Chemical Weapons Convention Chemicals Analysis: Sample Collection, Preparation, and Analytical Methods. Chichester, West Sussex, England: John Wiley & Sons; 2005: 403–431.

80. noort D, Black RM. Methods for the retrospective detection of exposure to toxic scheduled chemicals. Part B: mass spectrometric and immunochemical analysis of covalent adducts to proteins and DNA. In: Mesilaakso M, ed. Chemical Weapons Convention Chemicals Analysis: Sample Collection, Preparation, and Analytical Methods. Chichester, West Sussex, England: John Wiley & Sons; 2005: 433–451.

745 Medical Aspects of Chemical Warfare

81. Vycudilik W. Detection of mustard gas bis(2-chloroethyl)-sulfide in urine. Forensic Sci Int. 1985;28:131–136.

82. Vycudilik W. Detection of bis(2-chloroethyl)-sulfide (Yperite) in urine by high resolution gas chromatography-mass spectrometry. Forensic Sci Int. 1987;35:67–71.

83. Wils ERJ, Hulst AG, de Jong AL, Verweij A, Boter HL. Analysis of thiodiglycol in urine of victims of an alleged attack with mustard gas. J Anal Toxicol. 1985;9:254–257.

84. Wils ERJ, Hulst AG, van Laar J. Analysis of thiodiglycol in urine of victims of an alleged attack with mustard gas, part II. J Anal Toxicol. 1988;12:15–19.

85. drasch G, Kretschmer E, Kauert G, von Meyer L. Concentrations of mustard gas [bis(2-chloroethyl)sulfide] in the tissues of a victim of a vesicant exposure. J Forensic Sci. 1987;32:1788–1793.

86. Black RM, Read RW. Detection of trace levels of thiodiglycol in blood, plasma and urine using gas chromatography- electron-capture negative-ion chemical ionisation mass spectrometry. J Chromatogr. 1988;449: 261–270.

87. Jakubowski EM, Woodard CL, Mershon MM, Dolzine TW. Quantification of thiodiglycol in urine by electron ionization gas chromatography-mass spectrometry. J Chromatogr. 1990;528:184–190.

88. Black RM, Read RW. Improved methodology for the detection and quantitation of urinary metabolites of sulphur mustard using gas chromatography-tandem mass spectrometry. J Chromatogr B Biomed Appl. 1995;665:97–105.

89. Boyer AE, Ash D, Barr DB, et al. Quantitation of the sulfur mustard metabolites 1,1’-sulfonylbis[2-(methylthio)ethane] and thiodiglycol in urine using isotope-dilution gas chromatography-tandem mass spectrometry. J Anal Toxicol. 2004;28:327–332.

90. Black RM, Read RW. Methods for the analysis of thiodiglycol sulphoxide, a metabolite of sulphur mustard, in urine using gas chromatography-mass spectrometry. J Chromatogr. 1991;558:393–404.

91. Black RM, Read RW. Biological fate of sulphur mustard, 1,1’-thiobis(2-chloroethane): identification of beta-lyase metabolites and hydrolysis products in human urine. Xenobiotica. 1995;25:167–173.

92. Black RM, Brewster K, Clarke RJ, Hambrook JL, Harrison JM, Howells DJ. Biological fate of sulphur mustard, 1,1’- thiobis(2-chloroethane): isolation and identification of urinary metabolites following intraperitoneal administration to rat. Xenobiotica. 1992;22:405–418.

93. Black RM, Hambrook JL, Howells DJ, Read RW. Biological fate of sulfur mustard, 1,1’-thiobis(2-chloroethane). Urinary excretion profiles of hydrolysis products and beta-lyase metabolites of sulfur mustard after cutaneous application in rats. J Anal Toxicol. 1992;16:79–84.

94. Black RM, Clarke RJ, Read RW. Analysis of 1,1’-sulphonylbis[2-(methylsulphinyl)ethane] and 1-methylsulphinyl- 2-[2-(methylthio)ethylsulphonyl]ethane, metabolites of sulphur mustard, in urine using gas chromatography-mass spectrometry. J Chromatogr. 1991;558:405–414.

95. young CL, Ash D, Driskell WJ, et al. A rapid, sensitive method for the quantitation of specific metabolites of sulfur mustard in human urine using isotope-dilution gas chromatography-tandem mass spectrometry. J Anal Toxicol. 2004;28:339–345.

96. Read RW, Black RM. Analysis of beta-lyase metabolites of sulfur mustard in urine by electrospray liquid chromatography-tandem mass spectrometry. J Anal Toxicol. 2004;28:346–351.

97. Read RW, Black RM. Analysis of the sulfur mustard metabolite 1,1’-sulfonylbis[2-S-(N-acetylcysteinyl)ethane] in urine by negative ion electrospray liquid chromatography-tandem mass spectrometry. J Anal Toxicol. 2004;28:352–356.

98. fidder A, Noort D, de Jong LP, Benschop HP, Hulst AG. N7-(2-hydroxyethylthioethyl)-guanine: a novel urinary metabolite following exposure to sulphur mustard. Arch Toxicol. 1996;70:854–855.

746 Medical Diagnostics

99. Sandelowsky I, Simon GA, Bel P, Barak R, Vincze A. N1-(2-hydroxyethylthioethyl)-4-methyl imidazole (4-met-1-imid- thiodiglycol) in plasma and urine: a novel metabolite following dermal exposure to sulphur mustard. Arch Toxicol. 1992;66:296–297.

100. price EO, Smith JR, Clark CR, Schlager JJ, Shih ML. MALDI-ToF/MS as a diagnostic tool for the confirmation sulfur mustard exposure. J Appl Toxicol. 2000;20(suppl 1):S193–S197.

101. Willems JL. Clinical management of mustard gas casualties. Ann Med Milit (Belgium). 1989;3(suppl):1–61.

102. Jakubowski EM, Sidell FR, Evans RA, et al. Quantification of thiodiglycol in human urine after an accidental sulfur mustard exposure. Toxicol Methods. 2000;10:143–150.

103. carroll LS. Sulfur mustard: cutaneous exposure. Clin Toxicol (Phila). 2005;43(1):55.

104. newmark J, Langer JM, Capacio B, Barr J, McIntosh RG. Liquid sulfur mustard exposure. Mil Med. 2007;172(2):196– 198.

105. Barr JR, Young CL, Woolfit AR, et al. Comprehensive quantitative tandem MS analysis of urinary metabolites and albumin adducts following an accidental human exposure to sulfur mustard. In: Proceedings of the 53rd Conference of the American Society of Mass Spectrometry. San Antonio, Tex: June 5–9, 2005.

106. ludlum DB, Austin-Ritchie P, Hagopian M, Niu TQ, Yu D. Detection of sulfur mustard-induced DNA modifications. Chem Biol Interact. 1994;91:39–49.

107. Van der Schans GP, Scheffer AG, Mars-Groenendijk RH, Fidder A, Benschop HP, Baan RA. Immunochemical detection of adducts of sulfur mustard to DNA of calf thymus and human white blood cells. Chem Res Toxicol. 1994;7:408–413.

108. Van der Schans GP, Mars-Groenendijk R, de Jong LP, Benschop HP, Noort D. Standard operating procedure for immunoslotblot assay for analysis of DNA/sulfur mustard adducts in human blood and skin. J Anal Toxicol. 2004;28:316– 319.

109. fidder A, Moes GWH, Scheffer AG, et al. Synthesis, characterization, and quantitation of the major adducts formed between sulfur mustard and DNA of calf thymus and human blood. Chem Res Toxicol. 1994;7:199–204.

110. noort D, Verheij ER, Hulst AG, de Jong LP, Benschop HP. Characterization of sulfur mustard induced structural modifications in human hemoglobin by liquid chromatography-tandem mass spectrometry. Chem Res Toxicol. 1996;9:781–787.

111. Black RM, Harrison JM, Read RW. Biological fate of sulphur mustard: in vitro alkylation of human haemoglobin by sulphur mustard. Xenobiotica. 1997;27:11–32.

112. noort D, Hulst AG, Trap HC, de Jong LPA, Benschop HP. Synthesis and mass spectrometric identification of the major amino acid adducts formed between sulphur mustard and haemoglobin in human blood. Arch Toxicol. 1997;71:171– 178.

113. fidder A, Noort D, de Jong AL, Trap HC, de Jong LPA, Benschop HP. Monitoring of in vitro and in vivo exposure to sulfur mustard by GC/MS determination of the N-terminal valine adduct in hemoglobin after a modified Edman degradation. Chem Res Toxicol. 1996;9:788–792.

114. Black RM, Clarke RJ, Harrison JM, Read RW. Biological fate of sulphur mustard: identification of valine and histidine adducts in haemoglobin from casualties of sulphur mustard poisoning. Xenobiotica. 1997;27:499–512.

115. noort D, Hulst AG, de Jong LP, Benschop HP. Alkylation of human serum albumin by sulfur mustard in vitro and in vivo: mass spectrometric analysis of a cysteine adduct as a sensitive biomarker of exposure. Chem Res Toxicol. 1999;12:715–721.

116. noort D, Fidder A, Hulst AG, Woolfitt AR, Ash D, Barr JR. Retrospective detection of exposure to sulfur mustard: improvements on an assay for liquid chromatography-tandem mass spectrometry analysis of albumin-sulfur mustard adducts. J Anal Toxicol. 2004;28:333–338.

747 Medical Aspects of Chemical Warfare

117. noort D, Fidder A, Hulst AG, de Jong LP, Benschop HP. Diagnosis and dosimetry of exposure to sulfur mustard: development of a standard operating procedure for mass spectrometric analysis of haemoglobin adducts: exploratory research on albumin and keratin adducts. J Appl Toxicol. 2000;20(suppl 1):S187–S192.

118. capacio BR, Smith JR, DeLion MT, et al. Monitoring sulfur mustard exposure by gas chromatography-mass spectrometry analysis of thiodiglycol cleaved from blood proteins. J Anal Toxicol. 2004;28:306–310.

119. noort D, Fidder A, Benschop HP, de Jong LP, Smith JR. Procedure for monitoring exposure to sulfur mustard based on modified edman degradation of globin. J Anal Toxicol. 2004;28:311–315.

120. Benschop HP, van der Schans GP, Noort D, Fidder A, Mars-Groenendijk RH, de Jong LP. Verification of exposure to sulfur mustard in two casualties of the Iran-Iraq conflict. J Anal Toxicol. 1997;21:249–251.

121. Benschop HP, Noort D, van der Schans GP, de Jong LP. Diagnosis and dosimetry of exposure to sulfur mustard: development of standard operating procedures; further exploratory research on protein adducts. Rijswijk, The Netherlands: TNO Prins Maurits Laboratory; 2000. Final report DAMD17-97-2-7002, ADA381035.

122. Korte WD, Walker EM, Smith JR, et al. The determination of sulfur mustard exposure by analysis of blood protein adducts. Wehrmed Mschr. 2005;49:327.

123. united Nations Security Council. Report of the mission dispatched by the Secretary-General to investigate allegations of the use of chemical weapons in the conflict between the Islamic Republic of Iran and Iraq. New York, NY: UN; 1986. Report S/17911.

124. Van der Schans GP, Noort D, Mars-Groenendijk RH, et al. Immunochemical detection of sulfur mustard adducts with keratins in the stratum corneum of human skin. Chem Res Toxicol. 2002;15:21–25.

125. Smith JR, Logan TP, Szafraniec LL, Jakubowski EM. Spectroscopic characterization of the geminal isomer of Lewisite. Anal Lett. 1995;28:1541–1554.

126. Waters WA, Williams JH. Hydrolyses and derivatives of some vesicant arsenicals. J Chem Soc. 1950; 18–22.

127. Jakubowski EM, Smith JR, Logan TP, et al. Verification of Lewisite exposure: quantification of chlorovinylarsonous acid in biological samples. In: Proceedings of the 1993 Medical Defense Bioscience Review. Aberdeen Proving Ground, Md: US Army Medical Research Institute of Chemical Defense; 1993. AD A275667.

128. Wooten JV, Ashley DL, Calafat AM. Quantitation of 2-chlorovinylarsonous acid in human urine by automated solid-phase microextraction-gas chromatography-mass spectrometry. J Chromatogr B Analyt Technol Biomed Life Sci. 2002;772:147–153.

129. logan TP, Smith JR, Jakubowski EM, Nielson RE. Verification of lewisite exposure by the analysis of 2-chlorovinyl arsonous acid in urine. Toxicol Meth. 1999;9:275–284.

130. fidder A, Noort D, Hulst AG, de Jong LP, Benschop HP. Biomonitoring of exposure to lewisite based on adducts to haemoglobin. Arch Toxicol. 2000;74:207–214.

131. uS Department of Health and Human Services, Public Health Service, Agency for Toxic Substances and Disease Registry. Toxicological profile for cyanide. Atlanta, Ga: USDHHS; 1997.

132. hibbs CM. Cyanide and nitrate toxicoses of cattle. Vet Hum Toxicol. 1979;21:401–403.

133. conn EE. Introduction. In: Evered D, Harnett S, eds. Cyanide Compounds in Biology. Chichester, West Sussex, England: John Wiley & Sons; 1988: 1–2.

134. purser DA, Grimshaw P, Berrill KR. Intoxication of cyanide in fires: a study in monkeys using polyacrylonitrile. Arch Environ Health. 1984:39:394–400.

135. Blanc P, Hogan M, Mallin K, Hryhorczuk D, Hessl S, Bernard B. Cyanide intoxication among silver-reclaiming workers. JAMA. 1985;253:367–371.

748 Medical Diagnostics

136. izraeli S, Israeli A, Danon Y. Pharmacological treatment of cyanide poisoning. Harefuah. 1988;114:338–342.

137. Kirk RL, Stenhouse NS. Ability to smell solutions of potassium cyanide. Nature. 1953;171:698–699.

138. Warburg O. Inhibition of the action of prussic acid in living cells. Hoppe-Seyler’s Z Physiol Chem. 1911;76:331–346.

139. Keilin D. Cytochrome and respiratory enzymes. Proc R Soc Lond B Biol Sci. 1929;104:206–251.

140. diPalma JR. Noxious gases and vapers: I. Carbon monoxide, cyanide, methemoglobin, and sulfhemoglobin. In: Drills Pharmacology in Medicine. New York, NY: McGraw-Hill; 1971: 1189–1205.

141. Way JL. Cyanide intoxication and its mechanism of antagonism. Annu Rev Pharmacol Toxicol. 1984;24:451–481.

142. farooqui MY, Ahmed AE. Molecular interaction of acrylonitrile and potassium cyanide with rat blood. Chem Biol Interact. 1982;38:145–159.

143. Blakley RL, Coop IE. The metabolism and toxicity of cyanides and cyanogenic blycosides in sheep. NZ J Sci Technol. 1949;31A(3):1–16.

144. Wood JL, Cooley SL. Detoxification of cyanide by cystine. J Biol Chem. 1956;218:449–457.

145. Smith RP. Toxic responses of the blood. In: Casarett LJ, Amdur MO, Doull J, Klaassen CD, eds. Casarett & Doull’s Toxicology: The Basic Science of Poisons. 10th ed. McGraw-Hill; 1995: 335–354.

146. ansell M, Lewis FA. A review of cyanide concentrations found in human organs. A survey of literature concerning cyanide metabolism, ‘normal’, non-fatal, and fatal cyanide levels. J Forensic Med.1970;17:148–155.

147. Seto Y. Stability and spontaneous production of blood cyanide during heating. J Forensic Sci. 1996;41:465–468.

148. Ballantyne C. Artifacts in the definition of toxicity by cyanides and cyanogens. Fundam Appl Toxicol. 1983;3:400-408.

149. mathur BB, Krishna G. Toxicodynamics of phosgene. In: Somani S, ed. Chemical Warfare Agents. San Diego, Calif: Academic Press, Inc; 1992: 237–254.

150. malveda M. CEH Product review: phosgene. In: SRI International. Chemical Economics Handbook. Menlo Park, Calif: SRI International; 2003: 687.

151. Focused on phosgene. Chemistry Business. 2002;30–31.

152. Sartori M. The War Gases. New York, NY: D Van Nostrand; 1939.

153. parshall, GW. Trends in processing and manufacturing that will affect implementation of the Chemical Weapons Convention. Pure Appl Chem. 2002;74:2259–2263.

154. diller WF. Medical phosgene problems and their possible solution. J Occup Med. 1978;20:189–193.

155. uS National Research Council Committee on Toxicology. Emergency and Continuous Exposure Limits for Selected Airborne Contaminants. Vol 1. Washington, DC: National Academy Press; 1984.

156. Borak J, Diller WF. Phosgene exposure: mechanisms of injury and treatment strategies. J Occup Environ Med. 2001;43:110– 119.

157. diller WF. Pathogenesis of phosgene poisoning. Toxicol Ind Health. 1985;1:7–15.

158. National Institute for Occupational Health and Safety. Pocket Guide to Chemical Hazards. Washington, DC: NIOSH; 2004.

749 Medical Aspects of Chemical Warfare

159. cucinell SA. Review of the toxicity of long-term phosgene exposure. Arch Environ Health. 1974;28:272–275.

160. polednak AP. Mortality among men occupationally exposed to phosgene in 1943–1945. Environ Res. 1980;22:357– 367.

161. diller WF, Zante R. Dose-effect relations in the effects of phosgene on man and animals. Literature review. Zentralblatt fuer Arbeitsmedizin, Arbeitsschutz, Prophylaxe Ergonomie. 1982;32:360–4, 366–368.

162. pastore A, Federici G, Bertini E, Piemonte F. Analysis of glutathione: implication in redox and detoxification. Clin Chim Acta. 2003;333:19–39.

163. Richie JP Jr, Skowronski L, Abraham P, Leutzinger Y. Blood glutathione concentrations in a large-scale human study. Clin Chem. 1996;42:64–70.

164. Reed DJ, Fariss MW. Glutathione depletion and susceptibility. Pharmacol Rev. 1994;36(suppl 2):25–33.

165. pohl LR., Branchflower RV, Highest RJ, et al. The formation of diglutathionyl dithiocarbonate as a metabolite of chloroform, bromotrichloromethane, and carbon tetrachloride. Drug Metab Dispos. 1981;9:334–339.

166. fabrizi L, Taylor GW, Edwards RJ, Boobis AR. Adducts of the chloroform metabolite phosgene. Adv Exp Med Biol. 2001;500:129–132.

167. anders MW, Stevens JL, Sprague RW, Shaath, Z, Ahmed AE. Metabolism of haloforms to carbon monoxide. II. In vivo studies. Drug Metab Dispos. 1978:6:556–560.

168. Sciuto AM, Strickland PT, Kennedy TP, Gurtner GH. Postexposure treatment with aminophylline protects against phosgene-induced acute lung injury. Exp Lung Res. 1997;23:317–332.

169. Sciuto AM, Hurt HH. Therapeutic treatments of phosgene-induced lung injury. Inhal Toxicol. 2004;16:565–580.

170. Sciuto AM. Assessment of early acute lung injury in rodents exposed to phosgene. Arch Toxicol. 1998;72:283–288.

171. Sciuto AM, Cascio MB, Moran TS, Forster JS. The fate of antioxidant enzymes in bronchoalveolar lavage fluid over 7 days in mice with acute lung injury. Inhal Toxicol. 2003;15:675–685.

172. Kubic VL, Anders MW. Metabolism of carbon tetrachloride to phosgene. Life Sci. 1980;26:2151–2155.

173. Gwilt PR, Radick LE, Li XY, Whalen JJ, Leaf CD. Pharmacokinetics of 2-oxothiazolidine-4-carboxylate, a cysteine prodrug, and cysteine. J Clin Pharmacol. 1998;38:945–950.

174. pastore A, Massoud R, Motti C, et al. Fully automated assay for total homocysteine, cysteine, cysteinylglycine, glutathione, cysteamine and 2-mercaptopropionylglycine in plasma and urine. Clin Chem. 1998;44:825–832.

175. Boettcher B, Meister A. Synthesis of L-2-oxothiazolidine-4-carboxylic acid. Anal Biochem. 1984;138:449–450.

176. porta P, Aebi S, Summer K, Lauterburg BH. L-2-oxothiazolidine-4-carboxylic acid, a cysteine prodrug: pharmacokinetics and effects on thiols in plasma and lymphocytes in human. J Pharmacol Exp Ther. 1991;257:331–334.

177. Kling A. Acute pulmonary edema following exposure to war gases. Comptes Rendus De L’Academie des Sciences. 1933;197:1782–1784.

178. cowlen MS, Hewitt WR, Schroeder F. 2-hexanone potentiation of [14C]chloroform hepatotoxicity: covalent interaction of a reactive intermediate with rat liver phospholipid. Toxicol Appl Pharmacol. 1984;73:478–491.

179. de Biasi A, Sbraccia M, Keizer J, Testai E, Vittozzi L. The regioselective binding of CHCI3 reactive intermediates to microsomal phospholipids. Chem Biol Interact. 1992;85:229–242.

180. Guastadisegni C, Guidoni L, Balduzzi M, Viti V, Di Consiglio E, Vittozzi L. Characterization of a phospholipid adduct formed in Sprague Dawley rats by chloroform metabolism: NMR studies. J Biochem Mol Toxicol. 1998;12:93–102.

750 Medical Diagnostics

181. Guastadisegni C, Balduzzi M, Mancuso MT, Di Consiglio E. Liver mitochondria alterations in chloroform-treated Sprague-Dawley rats. J Toxicol Environ Health A. 1999;57:415–429.

182. di Consiglio E, De Angelis G, Testai E, Vittozzi L. Correlation of a specific mitochondrial phospholipid-phosgene adduct with chloroform acute toxicity. Toxicology. 2001;159:43–53.

183. Jugg B, Jenner J, Rice P. The effect of perfluoroisobutene and phosgene on rat lavage fluid surfactant phospholipids. Hum Exp Toxicol. 1999;18:659–668.

184. pereira MA, Chang LW, Ferguson JL, Couri D. Binding of chloroform to the cysteine of hemoglobin. Chem Biol Interact. 1984;51:115–124.

185. Sciuto AM, Stotts RR, Chittenden V, Choung E, Heflin MD. Changes in absorbance at 413 nm in plasma from three rodent species exposed to phosgene. Biochem Biophys Res Commun. 1996;226:906–911.

186. noort D, Hulst AG, Fidder A, van Gurp RA, de Jong LP, Benschop HP. In vitro adduct formation of phosgene with albumin and hemoglobin in human blood. Chem Res Toxicol. 2000;13:719–726.

187. Sciuto AM, Carpin LB, Moran TS, Forster JS. Chronological changes in electrolyte levels in arterial blood and bronchoalveolar lavage fluid in mice after exposure to an edemagenic gas. Inhal Toxicol. 2003;15:663–674.

188. currie WD, Hatch GE, Frosolono MF. Pulmonary alterations in rats due to acute phosgene inhalation. Fundam Appl Toxicol. 1987;8:107–114.

189. Byrd GD, Paule RC, Sander LC, Sniegoski LT, White E 5th, Bausum HT. Determination of 3-quinuclidinyl benzilate (QNB) and its major metabolites in urine by isotope dilution gas chromatography/mass spectrometry. J Anal Toxicol. 1992;16:182–187.

751 Medical Aspects of Chemical Warfare

752